Cooling system, cooling execution device, cooling device, cooling execution method, cooling method, program, cooling execution program, and cooling program

Abstract

A cooling execution device is provided with: a prediction unit that predicts a temperature change of a control device that controls automatic driving of a vehicle and is mounted on the vehicle; and a cooling execution unit that starts cooling of the control device on the basis of the temperature change predicted by the prediction unit. The present invention provides a cooling execution method executed by a computer, the cooling execution method comprising: a prediction stage of predicting a temperature change of a control device that controls automatic driving of a vehicle and is mounted on the vehicle; and a cooling execution stage of starting cooling of the control device on the basis of the temperature change predicted in the prediction stage.

Description

Technical Field

[0001] The present invention relates to a cooling system, a cooling execution device, a cooling device, a cooling execution method, a cooling method, a program, a cooling execution program and a cooling program. Background Art

[0002] Patent Document 1 describes a vehicle having an automatic driving function.

[0003] Prior art literature

[0004] Patent Literature

[0005] Patent document 1: Japanese Patent Application Publication No. 2022-035198. Summary of the invention

[0006] (Technical problem to be solved by the invention)

[0007] It is desired to realize a cooling system, a cooling execution device, a cooling device, a cooling execution method, a cooling method, a program, a cooling execution program, and a cooling program that can perform efficient cooling in response to temperature changes of a control device of a vehicle.

[0008] (Technical solutions for solving technical problems)

[0009] According to a technical solution of the present invention, a cooling execution device is provided. The cooling execution device may include a prediction unit, the prediction unit predicts the temperature change of a control device, the control device controls the automatic driving of a vehicle and is mounted on the vehicle. The cooling execution device may include a cooling execution unit, the cooling execution unit starts cooling the control device based on the temperature change predicted by the prediction unit.

[0010] In the cooling execution device, the cooling execution unit may start cooling the control device in response to a prediction that the temperature of the control device will become higher than a preset threshold value.

[0011] In any of the cooling execution devices described above, the prediction unit may predict the temperature change of the control device through AI. The cooling execution device may also include: a model storage unit that stores a learning model generated by machine learning, which takes the information obtained by the control device as input and the temperature change of the control device as output, wherein the machine learning takes the information obtained by the control device and the temperature change of the control device when the control device obtains the information as learning data; and an information acquisition unit that acquires the information obtained by the control device. The prediction unit may predict the temperature change of the control device by inputting the information obtained by the information acquisition unit into the learning model. The information acquisition unit may acquire sensor information obtained by the control device from a sensor mounted on the vehicle from the sensor or the control device. The information acquisition unit may acquire from the control device an analysis result after the control device analyzes a camera image obtained by a camera mounted on the vehicle. The information acquisition unit may acquire external information received by the control device from an external device from the external device or the control device. The information acquisition unit may acquire the traffic information of the road on which the vehicle is located, which is received by the control device from the external device, from the external device or the control device.

[0012] In any of the cooling execution devices described above, the cooling execution unit may start cooling the control device using one or more cooling units selected from a plurality of cooling units, corresponding to the temperature of the control device predicted by the prediction unit. The plurality of cooling units may include one or more air cooling units, one or more water cooling units, and a plurality of liquid nitrogen cooling units. The prediction unit may predict the temperature changes of each of the plurality of parts of the control device, and the cooling execution unit may start cooling the control device using a cooling unit selected from a plurality of cooling units for cooling the plurality of parts of the control device, respectively, based on the prediction result of the prediction unit. The control device may have a plurality of processing chips respectively arranged at different positions of the control device, and each of the plurality of cooling units may be arranged at a position corresponding to each of the plurality of processing chips.

[0013] According to one aspect of the present invention, there is provided a program for causing a computer to function as the cooling execution device.

[0014] According to a technical solution of the present invention, a cooling execution method is provided, which is executed by a computer. The cooling execution method may include a prediction stage: predicting the temperature change of a control device, the control device controls the automatic driving of a vehicle and is mounted on the vehicle. The cooling execution method may include a cooling execution stage: starting the cooling of the control device based on the temperature change predicted in the prediction stage.

[0015] According to the technical solution, a cooling system is provided. The cooling system may include a server and a cooling execution device, wherein the server includes: an information acquisition unit that acquires position information of a vehicle having the cooling execution device and temperature information of a control device of the vehicle from the cooling execution device; and a prediction unit that predicts a temperature change of the control device based on the position information of the vehicle having the cooling execution device and the temperature information of the control device of the vehicle acquired by the information acquisition unit, and the cooling execution device includes: an information acquisition unit that acquires a prediction result of the temperature change of the control device predicted by the prediction unit of the server from the server; and a cooling execution unit that starts cooling of the control device based on the prediction result of the temperature change of the control device acquired by the information acquisition unit.

[0016] It may be that the server further includes a creation unit, which creates a mapping diagram representing the relationship between the position of the vehicle and the temperature of the control device based on the position information of the vehicle having the cooling execution device acquired by the information acquisition unit and the temperature information of the control device of the vehicle, and the prediction unit predicts the temperature change of the control device based on the position information of the vehicle having the cooling execution device acquired by the information acquisition unit, the temperature information of the control device of the vehicle, and the mapping diagram representing the relationship between the position of the vehicle and the temperature of the control device created by the creation unit.

[0017] It may be that the information acquisition unit of the server acquires the position information of the vehicle having the cooling execution device, the temperature information of the control device of the vehicle, and the information related to the type of the vehicle from the cooling execution device; the creation unit creates a mapping diagram representing the relationship between the position of the vehicle within the type of the vehicle and the temperature of the control device based on the position information of the vehicle having the cooling execution device, the temperature information of the control device of the vehicle, and the information related to the type of the vehicle acquired by the information acquisition unit; the prediction unit of the server predicts the temperature change of the control device based on the position information of the vehicle having the cooling execution device acquired by the information acquisition unit, the temperature information of the control device of the vehicle, and the mapping diagram representing the relationship between the position of the vehicle within the type of the vehicle and the temperature of the control device created by the creation unit.

[0018] The cooling execution unit may start cooling the control device from a predetermined time before the control device starts generating heat, using the prediction result of the temperature change of the control device acquired by the information acquisition unit.

[0019] The cooling execution unit may use the prediction result of the temperature change of the control device acquired by the information acquisition unit to start cooling the control device using one or more cooling units selected from a plurality of cooling units in accordance with the temperature change of the control device.

[0020] The plurality of cooling units may include one or more air cooling units, one or more water cooling units, and a plurality of liquid nitrogen cooling units.

[0021] According to the technical solution, a cooling execution method is provided, which is executed by a server and a cooling execution device. The cooling execution method may include: an information acquisition process, in which the server acquires the position information of the vehicle having the cooling execution device and the temperature information of the control device of the vehicle from the cooling execution device. The cooling execution method may include: a prediction process, in which the server predicts the temperature change of the control device based on the position information of the vehicle having the cooling execution device and the temperature information of the control device of the vehicle acquired by the information acquisition process. The cooling execution method may include: an information acquisition process, in which the cooling execution device acquires from the server the prediction result of the temperature change of the control device predicted by the prediction process of the server. The cooling execution method may include: a cooling execution process, in which the cooling execution device starts cooling the control device based on the prediction of the temperature change of the control device acquired by the information acquisition process.

[0022] According to the technical solution, a cooling program is provided, comprising a program executed by a server and a program executed by a cooling execution device. The cooling program causes the server to execute: an information acquisition step, acquiring the position information of a vehicle having the cooling execution device and the temperature information of a control device of the vehicle from the cooling execution device; and a prediction step, predicting the temperature change of the control device based on the position information of the vehicle having the cooling execution device and the temperature information of the control device of the vehicle acquired by the information acquisition step, and the cooling program causes the cooling execution device to execute: an information acquisition step, acquiring from the server the predicted result of the temperature change of the control device predicted by the prediction step of the server; and a cooling execution step, starting the cooling of the control device based on the predicted result of the temperature change of the control device acquired by the information acquisition step.

[0023] According to the technical solution, a cooling system is provided. The cooling system may include a server and a cooling execution device, wherein the server includes: an information acquisition unit that acquires position information of a vehicle having the cooling execution device, temperature information of a control device of the vehicle, information related to traffic conditions, and weather information from the cooling execution device; and a prediction unit that predicts temperature changes of the control device based on the position information of the vehicle having the cooling execution device, temperature information of the control device of the vehicle, information related to traffic conditions, and weather information acquired by the information acquisition unit, wherein the cooling execution device includes: an information acquisition unit that acquires a prediction result of the temperature change of the control device predicted by the prediction unit of the server from the server; and a cooling execution unit that starts cooling of the control device based on the prediction result of the temperature change of the control device acquired by the information acquisition unit.

[0024] It may be that the server further includes a creation unit, which creates a mapping diagram representing the relationship between the position of the vehicle, the information related to traffic conditions, the weather information, and the temperature of the control device based on the position information of the vehicle having the cooling execution device, the temperature information of the control device of the vehicle, the information related to traffic conditions, and the weather information acquired by the information acquisition unit, and the prediction unit predicts the temperature change of the control device based on the position information of the vehicle having the cooling execution device, the temperature information of the control device of the vehicle, the information related to traffic conditions, the weather information, and the mapping diagram representing the relationship between the position of the vehicle, the information related to traffic conditions, the weather information, and the temperature of the control device created by the creation unit.

[0025] The creation unit can create a mapping diagram representing the relationship between the position of the vehicle and the temperature of the control device according to each of the information related to the traffic condition and the weather information, based on the position information of the vehicle having the cooling execution device, the temperature information of the control device of the vehicle, the information related to the traffic condition, and the weather information acquired by the information acquisition unit.

[0026] It may be that the information acquisition unit of the server also acquires information related to the type of the vehicle from the cooling execution device, and the creation unit creates a mapping diagram representing the relationship between the position of the vehicle, the information related to the traffic condition, the weather information, the type of the vehicle, and the temperature of the control device based on the position information of the vehicle having the cooling execution device, the temperature information of the control device possessed by the vehicle, the information related to the traffic condition, the weather information, and the information related to the type of the vehicle.

[0027] The cooling execution unit may start cooling the control device from a predetermined time before the control device starts generating heat, using the prediction result of the temperature change of the control device acquired by the information acquisition unit.

[0028] The cooling execution unit may use the prediction result of the temperature change of the control device acquired by the information acquisition unit to start cooling the control device using one or more cooling units selected from a plurality of cooling units in accordance with the temperature change of the control device.

[0029] The plurality of cooling units may include one or more air cooling units, one or more water cooling units, and a plurality of liquid nitrogen cooling units.

[0030] According to the technical solution, a cooling execution method is provided, which is executed by a server and a cooling execution device. The cooling execution method may include: an information acquisition process, in which the server acquires the position information of the vehicle having the cooling execution device, the temperature information of the control device of the vehicle, the information related to the traffic conditions, and the weather information from the cooling execution device. It may include: a prediction process, in which the server predicts the temperature change of the control device based on the position information of the vehicle having the cooling execution device, the temperature information of the control device of the vehicle, the information related to the traffic conditions, and the weather information acquired by the information acquisition process. It may include: an information acquisition process, in which the cooling execution device acquires from the server the predicted result of the temperature change of the control device predicted by the prediction process of the server. It may include: a cooling execution process, in which the cooling execution device starts cooling the control device based on the prediction of the temperature change of the control device acquired by the information acquisition process.

[0031] According to the technical solution, a cooling program is provided, which includes a program executed by a server and a program executed by a cooling execution device. The program may be that the server executes: an information acquisition step, acquiring the position information of the vehicle having the cooling execution device and the temperature information of the control device of the vehicle from the cooling execution device; and a prediction step, predicting the temperature change of the control device based on the position information of the vehicle having the cooling execution device, the temperature information of the control device of the vehicle, the information related to the traffic conditions, and the weather information acquired by the information acquisition step, and the program causes the cooling execution device to execute: an information acquisition step, acquiring from the server the predicted result of the temperature change of the control device predicted by the prediction step of the server; and a cooling execution step, starting the cooling of the control device based on the predicted result of the temperature change of the control device acquired by the information acquisition step.

[0032] According to the technical solution, a cooling system is provided. It may be that the cooling system includes a server and a cooling execution device, the server includes: an information acquisition unit, which acquires the position information of each integrated circuit in the control device, the information related to the processing of each integrated circuit, and the information related to the driving condition of the vehicle from the cooling execution device; and a prediction unit, which predicts the temperature change of each position of the control device based on the position information of each integrated circuit in the control device, the information related to the processing of each integrated circuit, and the information related to the driving condition of the vehicle acquired by the information acquisition unit, and the cooling execution device includes: an information acquisition unit, which acquires the prediction result of the temperature change of each position of the control device predicted by the prediction unit of the server from the server; and a cooling execution unit, which starts cooling the control device based on the prediction result of the temperature change of each position of the control device acquired by the information acquisition unit.

[0033] The prediction unit can predict the temperature changes over time at various positions of the control device based on the position information of each integrated circuit in the control device acquired by the information acquisition unit, the information related to the processing of each integrated circuit, and the information related to the driving condition of the vehicle.

[0034] The cooling execution unit may start cooling of a position where heat is generated in the control device using the prediction result of the temperature change at each position of the control device acquired by the information acquisition unit.

[0035] The cooling execution unit may start cooling the position where the control device generates heat from a predetermined time before the control device starts generating heat, using the prediction result of the temperature change at each position of the control device acquired by the information acquisition unit.

[0036] The cooling execution unit can use the predicted results of the temperature changes at various positions of the control device acquired by the information acquisition unit to start cooling the control device using one or more cooling units selected from a plurality of cooling units in response to the temperature changes at various positions of the control device.

[0037] The plurality of cooling units may include one or more air cooling units, one or more water cooling units, and a plurality of liquid nitrogen cooling units.

[0038] According to the technical solution, a cooling execution method is provided, which is executed by a server and a cooling execution device. It may include: an information acquisition process, in which the server acquires the position information of each integrated circuit in the control device, the information related to the processing of each integrated circuit, and the information related to the driving condition of the vehicle from the cooling execution device. It may include: a prediction process, in which the server predicts the temperature change of each position of the control device based on the position information of each integrated circuit in the control device, the information related to the processing of each integrated circuit, and the information related to the driving condition of the vehicle acquired by the information acquisition process. It may include: an information acquisition process, in which the cooling execution device acquires from the server the predicted result of the temperature change of each position of the control device predicted by the prediction process of the server. It may include: a cooling execution process, in which the cooling execution device starts cooling the control device based on the predicted result of the temperature change of each position of the control device acquired by the information acquisition process.

[0039] According to the technical solution, a cooling program is provided, comprising a program executed by a server and a program executed by a cooling execution device. The program may be such that the server executes: an information acquisition step, acquiring position information of each integrated circuit in the control device, information related to the processing of each integrated circuit, and information related to the driving condition of the vehicle from the cooling execution device; and a prediction step, predicting the temperature change of each position of the control device based on the position information of each integrated circuit in the control device, information related to the processing of each integrated circuit, and information related to the driving condition of the vehicle acquired by the information acquisition step, and the program causes the cooling execution device to execute: an information acquisition step, acquiring from the server the predicted result of the temperature change of each position of the control device predicted by the prediction step of the server; and a cooling execution step, starting the cooling of the control device based on the predicted result of the temperature change of each position of the control device acquired by the information acquisition step.

[0040] According to the technical solution, a cooling execution device is provided. The cooling execution device may include: a detection unit that detects the temperature of a control device that controls the automatic driving of a vehicle and is mounted on the vehicle; and a cooling execution unit that uses a given cooling unit to execute cooling of the control device based on the time that the temperature detected by the detection unit continues to be above a given temperature.

[0041] In the cooling execution device, when the time for which the temperature remains above the given temperature exceeds a given threshold, the cooling execution unit may perform cooling using cooling units including one or more air cooling units, one or more water cooling units, and multiple types of liquid nitrogen cooling units.

[0042] In the cooling execution device, the cooling execution unit may execute rapid cooling when the temperature exceeds a given threshold value.

[0043] In the cooling execution device, there may be a prediction unit that predicts the temperature change of a control device that controls the automatic driving of a vehicle and is mounted on the vehicle. The cooling execution unit uses the cooling unit to execute cooling based on the time that the temperature continues to be above the given temperature as the temperature change predicted by the prediction unit.

[0044] The cooling execution device may further include a prediction unit, which predicts temperature changes of each of the plurality of parts of the control device. The cooling execution unit starts cooling the control device based on the prediction result of the prediction unit, using a cooling unit selected from a plurality of cooling units that cool the plurality of parts of the control device respectively.

[0045] The control device may include a plurality of processing chips respectively arranged at different positions of the control device, and each of the plurality of cooling units may be arranged at a position corresponding to each of the plurality of processing chips.

[0046] According to one aspect of the present invention, there is provided a program for causing a computer to function as a cooling execution device.

[0047] According to a technical solution of the present invention, there is provided a cooling execution method, which is executed by a computer. The cooling execution method may include: a detection phase, detecting the temperature of a control device, which controls the automatic driving of a vehicle and is mounted on the vehicle; and a cooling execution phase, based on the time that the temperature detected by the detection phase continues to be above a given temperature, using a given cooling unit to execute cooling of the control device.

[0048] According to the technical solution, a cooling execution device is provided. The cooling execution device may include: an estimation unit that estimates the switching between automatic driving and manual driving based on a preset driving route; a creation unit that creates a cooling plan that sets cooling conditions based on a given cooling unit based on the estimation result of the estimation unit; and a cooling execution unit that executes cooling of a control device based on the cooling plan created by the creation unit, the control device controlling the automatic driving of a vehicle and mounted on the vehicle.

[0049] In the cooling execution device, the estimating unit may estimate whether or not to switch from the automatic driving to the manual driving based on a road condition of the driving route.

[0050] In the cooling execution device, it may be that when the estimation unit estimates that the automatic driving is to be switched to the manual driving, the creation unit creates the cooling plan with the cooling unit set for the driving range of the manual driving, and when the estimation unit estimates that the manual driving is to be switched to the automatic driving, the creation unit creates the cooling plan with the cooling unit set for the driving range of the automatic driving.

[0051] In the cooling execution device, it may be that it also has: a detection unit, which detects the temperature change of the control device, and the control device controls the automatic driving of the vehicle and is installed in the vehicle; and a prediction unit, which predicts the temperature change of the control device, and the creation unit compares the actual temperature change detected by the detection unit with the predicted temperature change predicted by the prediction unit, and when the predicted temperature change is greater than the actual temperature change, creates the cooling plan set with the given cooling unit.

[0052] In the cooling execution device, the cooling unit may include one or more air cooling units, one or more water cooling units, and multiple types of liquid nitrogen cooling units.

[0053] In the cooling execution device, there may be a prediction unit that predicts temperature changes in a control device that controls automatic driving of a vehicle and is mounted on the vehicle. The cooling execution unit executes cooling of the control device based on the cooling plan created by the creation unit, using a cooling unit selected from a plurality of cooling units that cool a plurality of parts of the control device respectively.

[0054] The control device may include a plurality of processing chips respectively arranged at different positions of the control device, and each of the plurality of cooling units may be arranged at a position corresponding to each of the plurality of processing chips.

[0055] According to one aspect of the present invention, there is provided a program for causing a computer to function as a cooling execution device.

[0056] According to a technical solution of the present invention, there is provided a cooling execution method, which is executed by a computer, and the cooling execution method may include: an estimation stage, in which the switching between automatic driving and manual driving is estimated based on a pre-set driving route; a creation stage, in which a cooling plan with cooling conditions based on a given cooling unit is created based on the estimation result of the estimation stage; and a cooling execution stage, in which the cooling of a control device is executed based on the cooling plan created by the creation stage, and the control device controls the automatic driving of the vehicle and is mounted on the vehicle.

[0057] According to a technical solution of the present invention, there is provided a cooling device comprising: a prediction unit which predicts temperature changes of a control device which controls automatic driving of a vehicle and is mounted on the vehicle; and a cooling execution unit which performs cooling based on the temperature changes predicted by the prediction unit to keep the temperature of the control device within a given temperature range.

[0058] Furthermore, in the cooling device, the cooling execution unit may determine the predetermined temperature range based on a threshold temperature preset for the control device.

[0059] Furthermore, the prediction unit may predict the temperature of the control device through AI.

[0060] In addition, the cooling device may further include: a model storage unit that stores a learning model generated by machine learning, which uses the information obtained by the control device as input and the temperature change of the control device as output, wherein the machine learning uses the information obtained by the control device and the temperature change of the control device when the control device obtains the information as learning data; and an information acquisition unit that acquires the information obtained by the control device. The prediction unit may predict the temperature change of the control device by inputting the information obtained by the information acquisition unit into the learning model. The information acquisition unit may acquire sensor information obtained by the control device from a sensor mounted on the vehicle from the sensor or the control device.

[0061] Furthermore, the information acquisition unit may acquire, from the control device, an analysis result of an image captured by a camera mounted on the vehicle and analyzed by the control device.

[0062] In addition, the information acquisition unit may acquire external information received by the control device from an external device from the external device or the control device. The information acquisition unit may acquire traffic information of the road on which the vehicle is located received by the control device from the external device from the external device or the control device.

[0063] Furthermore, the cooling execution unit may start cooling the control device using one or more cooling units selected from a plurality of cooling units in accordance with the level of the temperature of the control device predicted by the prediction unit.

[0064] In addition, the multiple cooling units in the cooling execution unit may include one or more air cooling units, one or more water cooling units, and multiple types of liquid nitrogen cooling units.

[0065] Furthermore, the prediction unit may predict temperature changes of each of the plurality of parts of the control device, and the cooling execution unit may start cooling the control device based on the prediction result of the prediction unit using a cooling unit selected from a plurality of cooling units that cool the plurality of parts of the control device respectively.

[0066] In addition, the control device may include a plurality of processing chips respectively arranged at different positions of the control device, and each of the plurality of cooling units may be arranged at a position corresponding to each of the plurality of processing chips.

[0067] According to a technical solution of the present invention, there is provided a cooling method which is performed by a cooling device, the cooling method comprising: a prediction process for predicting temperature changes of a control device which controls automatic driving of a vehicle and is mounted on the vehicle; and a cooling execution process for performing cooling based on the temperature changes predicted by the prediction process to keep the temperature of the control device within a given temperature range.

[0068] According to a technical solution of the present invention, there is provided a cooling program for causing a computer to execute the following processes: a prediction process for predicting temperature changes of a control device which controls automatic driving of a vehicle and is mounted on the vehicle; and a cooling execution process for performing cooling to keep the temperature of the control device within a given temperature range based on the temperature changes predicted in the prediction process.

[0069] According to the technical solution, a cooling execution device is provided. The cooling execution device may include: a detection unit that detects a temperature change of a control device that controls the automatic driving of a vehicle and is mounted on the vehicle; a selection unit that selects a given operating condition that reduces the temperature of the control device when the temperature change exceeds a given threshold; and an output unit that outputs the given operating condition based on the selection result of the selection unit.

[0070] In the cooling execution device, when the temperature change exceeds a given threshold value, the selection unit may select, as the given operating condition, an operating condition that reduces the amount of calculation involved in given information processing by the control device.

[0071] The cooling execution device may further include an information acquisition unit, which acquires the result of information processing related to the autonomous driving from an external information processing device, and the external information processing device performs the information processing related to the autonomous driving as the given information processing of the control device.

[0072] In the cooling execution device, when the temperature change exceeds a given threshold value, the selection unit may select, as the given operating condition, an operating condition for suppressing the driving speed of the automatic driving to be equal to or lower than a given speed.

[0073] In the cooling execution device, when the temperature change exceeds a given threshold value, the selection unit may select an operating condition for changing the automatic driving to a manual driving as the given operating condition.

[0074] In the cooling execution device, when the temperature change exceeds a given threshold value, the selection unit may select, as the given operating condition, an operating condition of stopping at a given position of a road included in the autonomous driving driving route.

[0075] In the cooling execution device, when the temperature change exceeds a given threshold value, the selection unit may select, as the given operating condition, an operating condition that suppresses acquisition of given information used in information processing for the autonomous driving.

[0076] According to one aspect of the present invention, there is provided a program for causing a computer to function as a cooling execution device.

[0077] According to a technical solution of the present invention, there is provided a cooling execution method, which is executed by a computer, and the cooling execution method comprises: a detection phase, detecting a temperature change of a control device, wherein the control device controls the automatic driving of a vehicle and is mounted on the vehicle; a selection phase, selecting a given operating condition that reduces the temperature of the control device when the temperature change exceeds a given threshold; and an output phase, outputting the given operating condition based on the selection result of the selection phase.

[0078] According to a technical solution of the present invention, there is provided a cooling device comprising: a prediction unit which predicts temperature changes of a control unit which controls automatic driving of a vehicle and is mounted on the vehicle; a communication unit which communicates with other vehicles existing within a given range based on the temperature changes predicted by the prediction unit; and an instruction unit which instructs the control unit to perform operations related to the control of automatic driving of other vehicles with which the communication unit communicates.

[0079] Furthermore, the communication unit may determine that another vehicle existing within a range in which communication with another vehicle can be maintained for a given time is a vehicle existing within the given range and perform communication.

[0080] Furthermore, the cooling device may further include a cooling execution unit configured to start cooling of the control device based on the temperature change predicted by the prediction unit.

[0081] Furthermore, the cooling execution unit may start cooling the control device in response to the prediction by the prediction unit that the temperature of the control device will become higher than a preset threshold value.

[0082] Furthermore, the prediction unit may predict the temperature change of the control unit through AI.

[0083] In addition, the automatic driving control device may also include: a model storage unit, which stores a learning model generated by machine learning, which takes the information obtained by the control device as input and the temperature change of the control device as output, and the machine learning uses the information obtained by the control device and the temperature change of the control device when the control device obtains the information as learning data; and an information acquisition unit, which acquires the information obtained by the control device, and the prediction unit can predict the temperature change of the control device by inputting the information obtained by the information acquisition unit into the learning model.

[0084] Furthermore, the information acquisition unit may acquire sensor information acquired by the control device from a sensor mounted on the vehicle from the sensor or the control device.

[0085] In addition, the information acquisition unit may acquire, from the control device, an analysis result of a camera image analyzed by the control device, the camera image being captured by a camera mounted on the vehicle.

[0086] Furthermore, the information acquisition unit may acquire external information received by the control device from an external device from the external device or the control device. The information acquisition unit may acquire traffic information of the road where the vehicle is located received by the control device from the external device from the external device or the control device.

[0087] Furthermore, the cooling execution unit may start cooling the control device using one or more cooling units selected from a plurality of cooling units, depending on the level of the temperature of the control device predicted by the prediction unit.

[0088] Furthermore, the plurality of cooling units in the cooling execution unit may include one or more air cooling units, one or more water cooling units, and a plurality of liquid nitrogen cooling units.

[0089] In addition, the prediction unit can predict the temperature changes of each of the multiple parts of the control device, and the cooling execution unit can start cooling the control device based on the prediction result of the prediction unit using a cooling unit selected from a plurality of cooling units that cool the multiple parts of the control device respectively.

[0090] Furthermore, the control device may include a plurality of processing chips respectively arranged at different positions of the control device, and each of the plurality of cooling units may be arranged at a position corresponding to each of the plurality of processing chips.

[0091] According to a technical solution of the present invention, there is provided a cooling method, which is performed by a cooling device, and the cooling method includes: a prediction process, which predicts the temperature change of a control device, which controls the automatic driving of a vehicle and is mounted on the vehicle; a communication process, which communicates with other vehicles existing in a given range based on the temperature change predicted by the prediction process; and an instruction process, which instructs the control device to perform operations related to the control of the automatic driving of other vehicles communicated with through the communication process.

[0092] According to a technical solution of the present invention, a cooling program is provided for causing a computer to execute the following processes: a prediction process for predicting temperature changes of a control device, which controls the automatic driving of a vehicle and is mounted on the vehicle; a communication process for communicating with other vehicles existing within a given range based on the temperature changes predicted by the prediction process; and an instruction process for instructing the control device to perform operations related to the control of the automatic driving of other vehicles communicated with through the communication process.

[0093] According to a technical solution of the present invention, a cooling execution device is provided. The cooling execution device may include: a prediction unit that predicts changes in the computing power of a control device, the control device controls the automatic driving of a vehicle and is mounted on the vehicle. The cooling execution device may include: a cooling execution unit that starts cooling the control device based on the changes predicted by the prediction unit.

[0094] In the cooling execution device, the prediction unit can predict the change of the computing power of the control device through AI. The prediction unit can predict the moment when the computing power of the control device becomes maximum, and the cooling execution unit can start cooling the control device based on the moment predicted by the prediction unit. The cooling execution unit can start cooling the control device at the moment when the computing power of the control device becomes maximum. The prediction unit can predict the time until the temperature of the control device becomes higher than a preset temperature based on the prediction result of the moment when the computing power of the control device becomes maximum, and the cooling execution unit can start cooling the control device based on the time predicted by the prediction unit.

[0095] Any of the cooling execution devices described above may further include: a model storage unit that stores a learning model generated by machine learning, which uses the information obtained by the control device as input and the computing power of the control device as output, wherein the machine learning uses the information obtained by the control device and the computing power of the control device when the control device obtains the information as learning data; and an information acquisition unit that acquires the information obtained by the control device, and the prediction unit can predict the change in the computing power of the control device by inputting the information acquired by the information acquisition unit into the learning model. The information acquisition unit can acquire at least any one of sensor information acquired by the control device from a sensor mounted on the vehicle, an analysis result of the control device analyzing a camera image captured by a camera mounted on the vehicle, external information received by the control device from an external device, and traffic information on the road where the vehicle is located received by the control device from the external device.

[0096] Any of the cooling execution devices described above may further include: a model storage unit, which stores a learning model generated by machine learning, which takes the information obtained by the control device as input and takes the time from when the cooling of the control device is started to when the control device is cooled as output, wherein the machine learning takes the information obtained by the control device and the time from when the cooling of the control device is started when the control device obtains the information as learning data; and an information acquisition unit, which acquires the information obtained by the control device, and the cooling execution unit may determine the timing to start cooling of the control device based on the changes predicted by the prediction unit and the time from when the cooling of the control device is started to when the control device is cooled, which is obtained by inputting the information obtained by the information acquisition unit into the learning model.

[0097] In any of the cooling execution devices, the cooling execution unit may adjust the degree of cooling of the control device according to a risk rate, wherein the risk rate indicates the probability of occurrence of a risk that the control device becomes hot and thus the control device no longer operates normally, thereby affecting normal automatic driving of the vehicle. The cooling execution unit may cool the control device with a first cooling intensity when the risk rate is lower than a first threshold, cool the control device with a second cooling intensity that is stronger than the first cooling intensity when the risk rate is higher than the first threshold and lower than a second threshold that is higher than the first threshold, and cool the control device with a third cooling intensity that is stronger than the second cooling intensity when the risk rate is higher than the second threshold.

[0098] According to one aspect of the present invention, there is provided a program for causing a computer to function as the cooling execution device.

[0099] According to a technical solution of the present invention, a cooling execution method is provided, which is executed by a computer. The cooling execution method may include: a prediction stage, predicting changes in the computing power of a control device, the control device controls the automatic driving of a vehicle and is mounted on the vehicle. The cooling execution method may include: a cooling execution stage, starting cooling of the control device based on the changes predicted in the prediction stage.

[0100] In addition, the above summary of the invention does not list all the features required by the present invention. In addition, sub-combinations of these feature groups can also constitute inventions. BRIEF DESCRIPTION OF THE DRAWINGS

[0101] Figure 1 An example of the system 10 is schematically shown.

[0102] Figure 2 This is an explanatory diagram for explaining the learning phase in the system 10 .

[0103] Figure 3 It is an explanatory diagram for explaining the cooling execution stage in the system 10 .

[0104] Figure 4 An example of the SoCBox 400 and the cooling unit 600 is schematically shown.

[0105] Figure 5 An example of the SoCBox 400 and the cooling unit 600 is schematically shown.

[0106] Figure 6 An example of the SoCBox 400 and the cooling unit 600 is schematically shown.

[0107] Figure 7 An example of the system 10 is schematically shown.

[0108] Figure 8 This is an explanatory diagram for explaining the learning phase in the system 10 .

[0109] Fig. 9 It is an explanatory diagram for explaining the cooling execution stage in the system 10 .

[0110] Fig.10 An example of the SoCBox 400 and the cooling unit 600 is schematically shown.

[0111] Fig.11 An example of the SoCBox 400 and the cooling unit 600 is schematically shown.

[0112] Fig.12 An example of the SoCBox 400 and the cooling unit 600 is schematically shown.

[0113] Fig.13 An example of the system 10 is schematically shown.

[0114] Fig.14 This is an explanatory diagram for explaining the learning phase in the system 10 .

[0115] Fig.15 It is an explanatory diagram for explaining the cooling execution stage in the system 10 .

[0116] Fig.16 An example of the SoCBox 400 and the cooling unit 600 is schematically shown.

[0117] Fig.17An example of the SoCBox 400 and the cooling unit 600 is schematically shown.

[0118] Fig.18 An example of the SoCBox 400 and the cooling unit 600 is schematically shown.

[0119] Fig.19 An example of the system 10 is schematically shown.

[0120] Fig. 20 This is an explanatory diagram for explaining the learning phase in the system 10 .

[0121] Fig.21 It is an explanatory diagram for explaining the cooling execution stage in the system 10 .

[0122] Fig. 22 An example of the SoCBox 400 and the cooling unit 600 is schematically shown.

[0123] Fig.23 An example of the SoCBox 400 and the cooling unit 600 is schematically shown.

[0124] Fig.24 An example of the SoCBox 400 and the cooling unit 600 is schematically shown.

[0125] Fig.25 An example of the system 10 is schematically shown.

[0126] Fig.26 This is an explanatory diagram for explaining the learning phase in the system 10 .

[0127] Fig. 27 It is an explanatory diagram for explaining the cooling execution stage in the system 10 .

[0128] Fig.28 An example of the SoCBox 400 and the cooling unit 600 is schematically shown.

[0129] Fig.29 An example of the SoCBox 400 and the cooling unit 600 is schematically shown.

[0130] Fig.30 An example of the SoCBox 400 and the cooling unit 600 is schematically shown.

[0131] Fig.31 An example of the system 10 is schematically shown.

[0132] Fig.32 This is an explanatory diagram for explaining the learning phase in the system 10 .

[0133] Fig.33 It is an explanatory diagram for explaining the cooling execution stage in the system 10 .

[0134] Fig.34 An example of the SoCBox 400 and the cooling unit 600 is schematically shown.

[0135] Fig.35 An example of the SoCBox 400 and the cooling unit 600 is schematically shown.

[0136] Fig.36 An example of the SoCBox 400 and the cooling unit 600 is schematically shown.

[0137] Fig.37 It is a diagram schematically showing an example of the system 10 .

[0138] Fig.38 This is an explanatory diagram for explaining the learning phase in the system 10 .

[0139] Fig.39 It is an explanatory diagram for explaining the cooling execution stage in the system 10 .

[0140] Fig.40 An example of the SoCBox 400 and the cooling unit 600 is schematically shown.

[0141] Fig.41 An example of the SoCBox 400 and the cooling unit 600 is schematically shown.

[0142] Fig.42 An example of the SoCBox 400 and the cooling unit 600 is schematically shown.

[0143] Fig.43 An example of the system 10 is schematically shown.

[0144] Fig.44 This is an explanatory diagram for explaining the learning phase in the system 10 .

[0145] Fig.45 It is an explanatory diagram for explaining the cooling execution stage in the system 10 .

[0146] Fig.46 An example of the SoCBox 400 and the cooling unit 600 is schematically shown.

[0147] Fig.47 An example of the SoCBox 400 and the cooling unit 600 is schematically shown.

[0148] Fig.48 An example of the SoCBox 400 and the cooling unit 600 is schematically shown.

[0149] Fig.49 It is a diagram schematically showing an example of the system 10 .

[0150] Fig.50 It is an explanatory diagram of a state in which communication between vehicles is performed in the present embodiment.

[0151] Fig.51 This is an explanatory diagram for explaining the learning phase in the system 10 .

[0152] Fig.52 It is an explanatory diagram for explaining the cooling execution stage in the system 10 .

[0153] Fig.53 An example of the SoCBox 400 and the cooling unit 600 is schematically shown.

[0154] Fig.54 An example of the SoCBox 400 and the cooling unit 600 is schematically shown.

[0155] Fig.55 An example of the SoCBox 400 and the cooling unit 600 is schematically shown.

[0156] Fig.56 An example of the system 10 is schematically shown.

[0157] Fig.57 This is an explanatory diagram for explaining the learning phase in the system 10 .

[0158] Fig.58 It is an explanatory diagram for explaining the cooling execution stage in the system 10 .

[0159] Fig.59 An example of the SoCBox 400 and the cooling unit 600 is schematically shown.

[0160] Fig.60 An example of the SoCBox 400 and the cooling unit 600 is schematically shown.

[0161] Fig.61 An example of the SoCBox 400 and the cooling unit 600 is schematically shown.

[0162] Fig.62 This is a diagram schematically showing an example of the hardware configuration of a computer 1200 that functions as the management server 100 , the SoCBox 400 , the cooling execution device 500 , or the cooling device 700 . DETAILED DESCRIPTION

[0163] Hereinafter, the present invention will be described by way of the embodiments of the invention, but the following embodiments do not limit the invention involved in the claims. In addition, the combination of features described in the embodiments is not necessarily all essential to the technical means of the invention.

[0164] <First embodiment>

[0165] When a SoC (System on Chip) for autonomous driving performs advanced computing, heat generation becomes a technical problem. Therefore, in this embodiment, a technology, namely Synchronized Burst Chilling, is provided that flexibly applies AI (Artificial Intelligence) to optimize the rapid cooling of the SoCBox.

[0166] The SoCBox will immediately become hot, making it difficult to perform advanced calculations in the vehicle (a technical issue for fully autonomous driving). AI predicts the heat dissipation of the SoCBox and performs cooling at the same time as the heat dissipation, thereby preventing the SoCBox from becoming hot and enabling advanced calculations in the vehicle. It is expected that it can be used to cool not only the SoCBox but also the battery, which may become a solution to the high temperature caused by rapid charging.

[0167] Figure 1 An example of the system 10 is schematically shown. The system 10 includes a management server 100 . The system 10 includes a SoCBox 400 . The system 10 includes a cooling execution device 500 . The system 10 includes a cooling unit 600 .

[0168] SoCBox400, cooling execution device 500 and cooling unit 600 are mounted on the vehicle. SoCBox400 uses the sensor values ​​of multiple sensors mounted on the vehicle to control the automatic driving of the vehicle. The automatic driving control of the vehicle consumes a very high processing load, so there is a situation where SoCBox400 becomes very hot. If the SoCBox400 is too hot, it is possible that the operation of SoCBox400 cannot be performed normally, or it may have a bad effect on the vehicle.

[0169] The cooling execution device 500 involved in this embodiment predicts the temperature change of SoCBox400 and starts cooling SoCBox400 based on the temperature change. For example, the cooling execution device 500 responds to the situation that SoCBox400 starts to heat up and immediately starts cooling SoCBox400. By starting cooling earlier than the start of heating or at the same time as the start of heating, it is possible to reliably prevent SoCBox400 from becoming high temperature. In addition, compared with the case where SoCBox400 is always cooled, the energy required for cooling can be reduced.

[0170] The cooling execution device 500 can predict the temperature change of the SoCBox 400 through AI. The learning of the temperature change of the SoCBox 400 can be performed by using the data collected by the vehicle 200. For example, the management server 100 collects data from the vehicle 200 and performs learning. The subject that performs learning is not limited to the management server 100, and it can also be other devices.

[0171] The vehicle 200 is equipped with a SoCBox 400 and a temperature sensor 40 for measuring the temperature of the SoCBox 400. The SoCBox 400 controls the automatic driving of the vehicle 200 using the sensor values ​​of the multiple sensors mounted on the vehicle 200 and the external information received from the various servers 30. The server 30 may be an example of an external device. As examples of the various servers 30, a server that provides traffic information, a server that provides weather information, etc. may be cited. The SoCBox 400 sends the sensor values, external information, etc. used in the control of the automatic driving, and the temperature change of the SoCBox 400 during the control to the management server 100.

[0172] The management server 100 performs learning using information received from one or more SoCBox400. The management server 100 generates a learning model that uses the information obtained by SoCBox400 as input and the temperature change of SoCBox400 as output by performing machine learning using information such as sensor values ​​obtained by SoCBox400, external information, and the temperature change of SoCBox400 when SoCBox400 obtains such information as learning data.

[0173] The vehicle 300 is a vehicle having a cooling function according to the present embodiment. The vehicle 300 is equipped with a SoCBox 400, a cooling execution device 500, and a cooling unit 600. The cooling execution device 500 can receive a learning model generated by the management server 100 from the management server 100 and store the learning model.

[0174] The cooling execution device 500 obtains the sensor values ​​of multiple sensors mounted on the vehicle 300 obtained by the SoCBox400, the external information received from multiple servers 30, or obtains it from the SoCBox400, and inputs the obtained information into the learning model to predict the temperature change of the SoCBox400.

[0175] When it is predicted that the SoC Box 400 starts to generate heat, or when it is predicted that the temperature of the SoC Box 400 will become higher than a preset threshold, the cooling execution device 500 starts cooling the SoC Box 400 using the cooling unit 600 .

[0176] SoCBox400, cooling execution device 500, management server 100, and server 30 can communicate via network 20. Network 20 can include a vehicle network. Network 20 can include the Internet. Network 20 can include a LAN (Local Area Network). Network 20 can include a mobile communication network. The mobile communication network can comply with any of the communication methods after the 5G (5th Generation) communication method, the LTE (Long Term Evolution) communication method, the 3G (3rd Generation) communication method, and the 6G (6th Generation) communication method.

[0177] Figure 2 2 is an explanatory diagram for explaining the learning stage in the system 10. Here, as sensors 210 mounted on the vehicle 200, a camera 211, a LiDAR (Light Detection And Ranging) 212, a millimeter wave sensor 213, an ultrasonic sensor 214, an IMU sensor 215, and a GNSS (Global Navigation Satellite System) sensor 216 are exemplified. The vehicle 200 does not necessarily have to be equipped with all of these sensors, and may not have some of them, or may have sensors other than these sensors.

[0178] The SoCBox 400 acquires sensor information from each sensor included in the sensor 210. In addition, the SoCBox 400 can perform communication via the network 20, and the SoCBox 400 receives external information from a plurality of servers 30 respectively via the network 20. Then, the SoCBox 400 performs automatic driving control of the vehicle 200 using the acquired information.

[0179] The temperature sensor 40 measures the temperature change of the SoCBox 400. The SoCBox 400 transmits the sensor information received from the sensor 210, the external information received from the server 30, and the temperature change measured by the temperature sensor 40 when the automatic driving control is performed by acquiring these information to the management server 100.

[0180] The management server 100 includes an information acquisition unit 102 , a model generation unit 104 , and a model provision unit 106 . The information acquisition unit 102 acquires various information. The management server 100 can receive information transmitted from the SoCBox 400 .

[0181] The model generation unit 104 generates a learning model by performing machine learning using the information acquired by the information acquisition unit 102. The model generation unit 104 generates a learning model that uses the information acquired by SoCBox400 as input and the temperature change of SoCBox400 as output by performing machine learning using the information acquired by SoCBox400 and the temperature change of SoCBox400 when SoCBox400 acquires the information as learning data.

[0182] The model providing unit 106 provides the learning model generated by the model generating unit 104. The model providing unit 106 may transmit the learning model to the cooling execution device 500 mounted on the vehicle 300.

[0183] The system 10 may be configured to predict the temperature change of each of the multiple parts of the SoCBox400. In this case, the vehicle 200 may be provided with multiple temperature sensors 40 that respectively measure the temperature change of each of the multiple parts of the SoCBox400. The SoCBox400 may send the sensor information received from the sensor 210, the external information received from the server 30, and the temperature change measured by the multiple temperature sensors 40 when the automatic driving control is performed after obtaining the information to the management server 100. The model generation unit 104 performs machine learning by using the information obtained by the SoCBox400 and the temperature change of each of the multiple parts of the SoCBox400 when the SoCBox400 obtains the information as learning data, thereby generating a learning model that uses the information obtained by the SoCBox400 as input and the temperature change of each of the multiple parts of the SoCBox400 as output.

[0184] Figure 3 This is an explanatory diagram for explaining the cooling execution stage in the system 10. Here, as sensors 310 mounted on the vehicle 300, a camera 311, a LiDAR 312, a millimeter wave sensor 313, an ultrasonic sensor 314, an IMU sensor 315, and a GNSS sensor 316 are exemplified. The vehicle 300 does not necessarily have to be equipped with all of these sensors, and may not have some of them, or may have sensors other than these sensors.

[0185] The cooling execution device 500 includes a model storage unit 502, an information acquisition unit 504, a prediction unit 506, and a cooling execution unit 508. The model storage unit 502 stores the learning model received from the management server 100. The information acquisition unit 504 acquires the information acquired by the SoCBox 400.

[0186] The information acquisition unit 504 acquires the sensor information acquired by the SoCBox400 from the sensor 310 from the sensor 310 or the SoCBox400. For example, the information acquisition unit 504 may receive the sensor information acquired by the SoCBox400 from the sensor 310 from the SoCBox400. The information acquisition unit 504 may also receive the same sensor information from the sensor 310 as the sensor information acquired by the SoCBox400 from the sensor 310. In this case, each of the sensors 310 may send the sensor information to the SoCBox400 and each of the cooling execution devices 500.

[0187] The information acquisition unit 504 acquires the external information acquired by the SoCBox400 from the server 30 from the server 30 or the SoCBox400. The information acquisition unit 504 may receive the external information received by the SoCBox400 from the server 30 from the SoCBox400. The information acquisition unit 504 may also receive the same external information from the server 30 as the external information received by the SoCBox400 from the server 30. In this case, the server 30 may send the external information to each of the SoCBox400 and the cooling execution device 500.

[0188] The prediction unit 506 predicts the temperature change of the SoCBox 400. The prediction unit 506 may predict the temperature change of the SoCBox 400 through AI. For example, the prediction unit 506 predicts the temperature change of the SoCBox 400 by inputting the information acquired by the information acquisition unit 504 into the learning model stored in the model storage unit 502.

[0189] The cooling execution unit 508 starts cooling the SoCBox 400 based on the temperature change of the SoCBox 400 predicted by the prediction unit 506. For example, the cooling execution unit 508 starts cooling the SoCBox 400 in response to the prediction unit 506 predicting that the SoCBox 400 starts to heat up. For example, the cooling execution unit 508 starts cooling the SoCBox 400 in response to the prediction unit 506 predicting that the temperature of the SoCBox 400 will become higher than a preset threshold.

[0190] The cooling execution unit 508 may use the cooling unit 600 to execute cooling of the SoCBox 400. The cooling unit 600 may use an air cooling unit to cool the SoCBox 400. The cooling unit 600 may use a water cooling unit to cool the SoCBox 400. The cooling unit 600 may use a liquid nitrogen cooling unit to cool the SoCBox 400.

[0191] The cooling unit 600 may include multiple cooling units. For example, the cooling unit 600 includes multiple air cooling units. For example, the cooling unit 600 includes multiple water cooling units. For example, the cooling unit 600 includes multiple liquid nitrogen cooling units. The cooling unit 600 may include multiple of one or more air cooling units, one or more water cooling units, and one or more liquid nitrogen cooling units.

[0192] The plurality of cooling units may be configured to cool different parts of the SoCBox 400. The prediction unit 506 may use the information acquired by the information acquisition unit 504 to predict the temperature change of each of the plurality of parts of the SoCBox 400. The cooling execution unit 508 may start cooling the SoCBox 400 using a cooling unit selected from the plurality of cooling units that cool the plurality of parts of the SoCBox 400, based on the prediction result of the prediction unit 506.

[0193] The cooling execution unit 508 can use a cooling unit corresponding to the temperature of the SoCBox400 predicted by the prediction unit 506 to execute cooling of the SoCBox400. For example, the higher the temperature of the SoCBox400, the more cooling units the cooling execution unit 508 uses to execute cooling of the SoCBox400. As a specific example, when it is predicted that the temperature of the SoCBox400 will exceed a first threshold, the cooling execution unit 508 uses one of the plurality of cooling units to start cooling. Even so, when the temperature of the SoCBox400 still rises and is predicted to exceed a second threshold, the number of cooling units used is increased.

[0194] The cooling execution unit 508 can use a more powerful cooling unit to cool the SoCBox400 when the temperature of the SoCBox400 is higher. For example, when it is predicted that the temperature of the SoCBox400 will exceed the first threshold, the cooling execution unit 508 starts cooling using the air cooling unit. When the temperature of the SoCBox400 still rises despite this and is predicted to exceed the second threshold, the cooling execution unit 508 starts cooling using the water cooling unit. When the temperature of the SoCBox400 still rises despite this and is predicted to exceed the third threshold, the cooling execution unit 508 starts cooling using the liquid nitrogen cooling unit.

[0195] The SoCBox 400 may have a plurality of processing chips, and the plurality of processing chips may be respectively arranged at different positions of the SoCBox 400. Each of the plurality of cooling units may be arranged at a position corresponding to each of the plurality of processing chips.

[0196] For example, when the number of processing chips used varies according to the control conditions of driving, cooling is performed using a cooling unit corresponding to the processing chips used, thereby enabling efficient cooling.

[0197] Figure 4 An example of the SoCBox 400 and the cooling unit 600 is schematically shown. Figure 4 The cooling unit 600 is illustrated as being composed of one cooling unit. When the cooling execution device 500 predicts that the SoCBox 400 starts to generate heat or that the temperature of the SoCBox 400 exceeds a preset threshold, the cooling unit 600 starts cooling, thereby cooling the entire SoCBox 400.

[0198] Figure 5 An example of the SoCBox 400 and the cooling unit 600 is schematically shown. Figure 5 The example shows a case where the cooling unit 600 is composed of a plurality of cooling units for cooling a plurality of parts of the SoCBox 400. The cooling execution device 500 predicts the temperature change of each of the plurality of parts of the SoCBox 400, and in response to the prediction that any part starts to generate heat or the temperature of any part will exceed a preset threshold, performs cooling using only the cooling unit corresponding to the part, thereby achieving efficient cooling.

[0199] Figure 6 An example of the SoCBox 400 and the cooling unit 600 is schematically shown. Figure 6 The case where the cooling unit 600 is composed of two types of cooling units is illustrated. The cooling execution device 500 predicts the temperature change of each of the multiple parts of the SoCBox400, and in response to the prediction that any part begins to heat up or the temperature of any part will exceed a preset threshold, performs cooling using only the cooling unit corresponding to the part, thereby achieving efficient cooling. In addition, the cooling execution device 500 increases the cooling units used as the temperature of the SoCBox400 increases, that is, in this example, first, one of the two cooling units is used for cooling, and when the temperature of the SoCBox400 further increases, the other cooling unit is also used for cooling, thereby making the energy used for cooling more efficient.

[0200] <Second embodiment>

[0201] When the SoC (System on Chip) for autonomous driving performs advanced computing, heat generation becomes a technical problem. Therefore, in this embodiment, a technology called Synchronized Burst Chilling is provided that flexibly applies AI (Artificial Intelligence) to optimize the rapid cooling of the SoCBox.

[0202] The SoCBox will immediately become hot, making it difficult to perform advanced calculations in the vehicle (a technical issue for fully autonomous driving). AI predicts the heat dissipation of the SoCBox and performs cooling at the same time as the heat dissipation, thereby preventing the SoCBox from becoming hot and enabling advanced calculations in the vehicle. It is expected that it can be used to cool not only the SoCBox but also the battery, which may become a solution to the high temperature caused by rapid charging.

[0203] Figure 7 An example of the system 10 is schematically shown. The system 10 includes a management server 100 . The system 10 includes a SoCBox 400 . The system 10 includes a cooling execution device 500 . The system 10 includes a cooling unit 600 .

[0204] SoCBox400, cooling execution device 500 and cooling unit 600 are mounted on the vehicle. SoCBox400 is a control device that uses the sensor values ​​of multiple sensors mounted on the vehicle to control the automatic driving of the vehicle. The automatic driving control of the vehicle consumes a very high processing load, so there is a situation where SoCBox400 becomes very hot. If the SoCBox400 is too hot, it is possible that the operation of SoCBox400 cannot be performed normally, or it may have a bad effect on the vehicle.

[0205] The system 10 involved in this embodiment predicts the temperature change of SoCBox400 based on the position information of the vehicle having the cooling execution device 500 and the temperature information of the SoCBox400 possessed by the vehicle, and starts cooling the SoCBox400 based on the temperature change. For example, the system 10 predicts heating in advance in places such as urban areas with many vehicles or pedestrians where the SoCBox400 is likely to heat up due to increased processing volume, and immediately starts cooling the SoCBox400. By starting cooling earlier than the start of heating or at the same time as the start of heating, it is possible to reliably prevent the SoCBox400 from becoming high temperature. In addition, compared with the case where the SoCBox400 is always cooled, the energy required for cooling can be reduced.

[0206] The system 10 can predict the temperature change of the SoCBox 400 through AI. The learning of the temperature change of the SoCBox 400 can be performed using data collected by the vehicle 200. For example, the management server 100 collects data from the vehicle 200 to perform learning. The subject that performs learning is not limited to the management server 100, and it can also be other devices.

[0207] The vehicle 200 is equipped with a SoCBox 400 and a temperature sensor 40 for measuring the temperature of the SoCBox 400. The SoCBox 400 controls the automatic driving of the vehicle 200 using the sensor values ​​of the multiple sensors mounted on the vehicle 200 and the external information received from the various servers 30. The server 30 may be an example of an external device. As examples of the various servers 30, a server that provides traffic information, a server that provides weather information, etc. may be cited. The SoCBox 400 sends the sensor values, external information, etc. used in the control of the automatic driving, and the temperature change of the SoCBox 400 during the control to the management server 100.

[0208] The management server 100 performs learning using information received from one or more SoCBox400. The management server 100 generates a learning model that uses the information obtained by SoCBox400 as input and the temperature change of SoCBox400 as output by performing machine learning using information such as sensor values ​​obtained by SoCBox400, external information, and the temperature change of SoCBox400 when SoCBox400 obtains such information as learning data.

[0209] The vehicle 300 is a vehicle having a cooling function according to the present embodiment. The vehicle 300 is equipped with a SoCBox 400, a cooling execution device 500, and a cooling unit 600. The cooling execution device 500 can receive and store a learning model generated by the management server 100 from the management server 100. In addition, the cooling execution device 500 can receive and store information related to the type of the vehicle 300 from the server 30. In addition, the information related to the type of the vehicle 300 can include information such as the model of the vehicle 300, information related to the model and the heat generation of the SoCBox 400, and parameters of each model. Here, the parameters of each model include information such as information on models where heat generation of the SoCBox 400 is prone to occur, information on models where heat generation of the SoCBox 400 is difficult to occur, and information on cooling units that can be mounted on each model.

[0210] The cooling execution device 500 obtains sensor values ​​of multiple sensors mounted on the vehicle 300 obtained by the SoCBox400, and external information received from multiple servers 30, or obtains it from the SoCBox400, and inputs the obtained information into the learning model, thereby predicting the temperature change of the SoCBox400. In addition, the cooling execution device 500 can perform the same processing as the prediction processing performed by the management server 100 described later.

[0211] When it is predicted that the SoC Box 400 starts to generate heat and the temperature of the SoC Box 400 will become higher than a preset threshold, the cooling execution device 500 starts cooling the SoC Box 400 using the cooling unit 600 .

[0212] SoCBox400, cooling execution device 500, management server 100, and server 30 can communicate via network 20. Network 20 can include a vehicle network. Network 20 can include the Internet. Network 20 can include a LAN (Local Area Network). Network 20 can include a mobile communication network. The mobile communication network can comply with any of the communication methods after 5G (5th Generation) communication method, LTE (Long Term Evolution) communication method, 3G (3rd Generation) communication method, and 6G (6th Generation) communication method.

[0213] Figure 8 2 is an explanatory diagram for explaining the learning stage in the system 10. Here, as the sensor 210 mounted on the vehicle 200, a camera 211, a LiDAR (Light Detection And Ranging) 212, a millimeter wave sensor 213, an ultrasonic sensor 214, an IMU sensor 215, a GNSS (Global Navigation Satellite System) sensor 216, and a GPS (Global Positioning System) sensor 217 are exemplified. The vehicle 200 does not necessarily have to be equipped with all of these sensors, and may not have some of them, or may have sensors other than these sensors.

[0214] The SoCBox 400 acquires sensor information from each sensor included in the sensor 210. In addition, the SoCBox 400 can perform communication via the network 20, and the SoCBox 400 receives external information from a plurality of servers 30 respectively via the network 20. Then, the SoCBox 400 performs automatic driving control of the vehicle 200 using the acquired information.

[0215] The temperature sensor 40 measures the temperature change of the SoCBox 400. The SoCBox 400 transmits the sensor information received from the sensor 210, the external information received from the server 30, and the temperature change measured by the temperature sensor 40 when the automatic driving control is performed by acquiring these information to the management server 100.

[0216] The management server 100 includes an information acquisition unit 102 , a model generation unit 104 , a model provision unit 106 , a creation unit 108 , and a prediction unit 110 .

[0217] The information acquisition unit 102 acquires various information. For example, the information acquisition unit 102 acquires the position information of the vehicle 300 having the cooling execution device 500 and the temperature information of the SoCBox400 possessed by the vehicle 300 from the cooling execution device 500. In addition, the information acquisition unit 102 acquires the position information of the vehicle 300 having the cooling execution device 500, the temperature information of the SoCBox400 possessed by the vehicle 300, and the information related to the type of the vehicle 300 from the cooling execution device 500. In addition, the position information may also be information acquired from a GNSS sensor or a GPS sensor. In addition, the position information may include information related to the driving path and destination of the vehicle 300. In addition, the information acquisition unit 102 may acquire various information based on a given timing or cycle. The management server 100 may receive information sent by the SoCBox400.

[0218] The model generation unit 104 generates a learning model by performing machine learning using the information acquired by the information acquisition unit 102. The model generation unit 104 generates a learning model that uses the information acquired by SoCBox400 as input and the temperature change of SoCBox400 as output by performing machine learning using the information acquired by SoCBox400 and the temperature change of SoCBox400 when SoCBox400 acquires the information as learning data.

[0219] The model providing unit 106 provides the learning model generated by the model generating unit 104. The model providing unit 106 may transmit the learning model to the cooling execution device 500 mounted on the vehicle 300.

[0220] The creation unit 108 creates a map representing the relationship between the position of the vehicle 300 and the temperature of the SoCBox 400 based on the position information of the vehicle 300 having the cooling execution device 500 acquired by the information acquisition unit 102 and the temperature information of the SoCBox 400 possessed by the vehicle 300. In addition, the creation unit 108 creates a map representing the relationship between the position of the vehicle 300 and the temperature of the SoCBox 400 for each type of the vehicle 300 based on the position information of the vehicle 300 having the cooling execution device 500 acquired by the information acquisition unit 102, the temperature information of the SoCBox 400 possessed by the vehicle 300, and the information related to the type of the vehicle 300.

[0221] The prediction unit 110 predicts the temperature change of the SoCBox 400 based on the position information of the vehicle 300 having the cooling execution device 500 and the temperature information of the SoCBox 400 of the vehicle 300 acquired by the information acquisition unit 102. At this time, the prediction unit 110 may also use the position information of the vehicle 300 having the cooling execution device 500 and the temperature information of the SoCBox 400 of the vehicle 300 acquired by the information acquisition unit 102 in the past to perform the prediction.

[0222] For example, the prediction unit 110 uses the accumulated position information of the vehicle 300 having the cooling execution device 500 and the temperature information of the SoCBox 400 of the vehicle 300 to predict the temperature change of the SoCBox 400 based on the position information of the vehicle having the cooling execution device 500 and the temperature information of the SoCBox 400 of the vehicle 300 acquired by the information acquisition unit 102. For example, the prediction unit 110 predicts the temperature change based on the data that the heat generation increases by about 20% in the city compared to the normal time in the information acquired in the past, when the position information of the vehicle 300 is the urban area, based on the data that the heat generation increases by about 20% in the city compared to the normal time.

[0223] In addition, the prediction unit 110 predicts the temperature change of the SoCBox 400 based on the position information of the vehicle 300 having the cooling execution device 500 acquired by the information acquisition unit 102, the temperature information of the SoCBox 400 possessed by the vehicle 300, and the map representing the relationship between the position of the vehicle 300 and the temperature of the SoCBox 400 created by the creation unit 108. In addition, the prediction unit 110 predicts the temperature change of the SoCBox 400 based on the position information of the vehicle 300 having the cooling execution device 500 acquired by the information acquisition unit 102, the temperature information of the SoCBox 400 possessed by the vehicle 300, and the map representing the relationship between the position of the vehicle 300 and the temperature of the SoCBox 400 in the type of the vehicle 300 created by the creation unit 108.

[0224] The temperature change of SoCBox400 predicted by the prediction unit 110 includes the temperature change of SoCBox400 of the vehicle 300 over time. For example, the prediction unit 110 may predict the temperature change of SoCBox400 of the vehicle 300 over time on the travel route from the current position of the vehicle 300 to the destination.

[0225] The system 10 may also be configured to predict the temperature change of each of the multiple parts of the SoCBox400. In this case, the vehicle 200 may be provided with multiple temperature sensors 40 that measure the temperature change of each of the multiple parts of the SoCBox400. The SoCBox400 may send the sensor information received from the sensor 210, the external information received from the server 30, and the temperature change measured by the multiple temperature sensors 40 when the automatic driving control is performed after obtaining the information to the management server 100. The model generation unit 104 performs machine learning by using the information obtained by the SoCBox400 and the temperature change of each of the multiple parts of the SoCBox400 when the SoCBox400 obtains the information as learning data, thereby generating a learning model that uses the information obtained by the SoCBox400 as input and the temperature change of each of the multiple parts of the SoCBox400 as output.

[0226] Fig. 9 3 is an explanatory diagram for explaining the cooling execution stage in the system 10. Here, as sensors 310 mounted on the vehicle 300, a camera 311, a LiDAR 312, a millimeter wave sensor 313, an ultrasonic sensor 314, an IMU sensor 315, a GNSS sensor 316, and a GPS sensor 317 are exemplified. The vehicle 300 does not necessarily have to be equipped with all of these sensors, and may not have some of them, or may have sensors other than these sensors.

[0227] The cooling execution device 500 includes a model storage unit 502, an information acquisition unit 504, a prediction unit 506, and a cooling execution unit 508. The model storage unit 502 stores the learning model received from the management server 100. The information acquisition unit 504 acquires the information acquired by the SoCBox 400.

[0228] The information acquisition unit 504 acquires the sensor information acquired by the SoCBox400 from the sensor 310 from the sensor 310 or the SoCBox400. For example, the information acquisition unit 504 may receive the sensor information acquired by the SoCBox400 from the sensor 310 from the SoCBox400. The information acquisition unit 504 may also receive the same sensor information from the sensor 310 as the sensor information acquired by the SoCBox400 from the sensor 310. In this case, each of the sensors 310 may send the sensor information to the SoCBox400 and each of the cooling execution devices 500.

[0229] The information acquisition unit 504 acquires the external information acquired by the SoCBox400 from the server 30 from the server 30 or the SoCBox400. The information acquisition unit 504 may receive the external information received by the SoCBox400 from the server 30 from the SoCBox400. The information acquisition unit 504 may also receive the same external information from the server 30 as the external information received by the SoCBox400 from the server 30. In this case, the server 30 may send the external information to each of the SoCBox400 and the cooling execution device 500.

[0230] The information acquisition unit 504 acquires, from the management server 100 , the prediction result of the temperature change of the SoCBox 400 predicted by the prediction unit 110 of the management server 100 .

[0231] The prediction unit 506 predicts the temperature change of the SoCBox400. The prediction unit 506 can predict the temperature change of the SoCBox400 through AI. For example, the prediction unit 506 predicts the temperature change of the SoCBox400 by inputting the information acquired by the information acquisition unit 504 into the learning model stored in the model storage unit 502. In addition, the prediction unit 506 can perform the same prediction processing as the prediction processing performed by the prediction unit 110 of the management server 100.

[0232] The cooling execution unit 508 starts cooling the SoCBox400 based on the prediction result of the temperature change of the SoCBox400 acquired by the information acquisition unit 504. For example, the cooling execution unit 508 uses the prediction result of the temperature change of the SoCBox400 acquired by the information acquisition unit 504 to start cooling the SoCBox400 from a given time before the SoCBox400 starts to generate heat. In addition, for example, the cooling execution unit 508 uses the prediction result of the temperature change of the SoCBox400 acquired by the information acquisition unit 504 to start cooling the SoCBox400 from a given time before the timing when the SoCBox400 is predicted to become a given temperature or above. In addition, for example, the cooling execution unit 508 uses the prediction result of the temperature change of the SoCBox400 acquired by the information acquisition unit 504 to start cooling the SoCBox400 when the location where the SoCBox400 is predicted to start generating heat and the current position of the vehicle 300 become within a given distance.

[0233] The cooling execution unit 508 starts cooling the SoCBox 400 based on the temperature change of the SoCBox 400 predicted by the prediction unit 506. For example, the cooling execution unit 508 starts cooling the SoCBox 400 in response to the prediction unit 506 predicting that the SoCBox 400 starts to heat up. For example, the cooling execution unit 508 starts cooling the SoCBox 400 in response to the prediction unit 506 predicting that the temperature of the SoCBox 400 will become higher than a preset threshold.

[0234] The cooling execution unit 508 uses the predicted result of the temperature change of the SoCBox400 obtained by the information acquisition unit 504, and uses one or more cooling units selected from a plurality of cooling units to start cooling the SoCBox400 according to the temperature change of the SoCBox400. The cooling execution unit 508 can use the cooling unit 600 to execute the cooling of the SoCBox400. The cooling unit 600 can use an air cooling unit to cool the SoCBox400. The cooling unit 600 can use a water cooling unit to cool the SoCBox400. The cooling unit 600 can use a liquid nitrogen cooling unit to cool the SoCBox400.

[0235] The cooling unit 600 may also include multiple cooling units. For example, the cooling unit 600 includes multiple air cooling units. For example, the cooling unit 600 includes multiple water cooling units. For example, the cooling unit 600 includes multiple liquid nitrogen cooling units. The cooling unit 600 may also include multiple of one or more air cooling units, one or more water cooling units, and one or more liquid nitrogen cooling units.

[0236] The plurality of cooling units may be configured to cool different parts of the SoCBox 400. The prediction unit 506 may use the information acquired by the information acquisition unit 504 to predict the temperature change of each of the plurality of parts of the SoCBox 400. The cooling execution unit 508 may start cooling the SoCBox 400 using a cooling unit selected from the plurality of cooling units that cool the plurality of parts of the SoCBox 400, based on the prediction result of the prediction unit 506.

[0237] The cooling execution unit 508 can use a cooling unit corresponding to the temperature of the SoCBox400 predicted by the prediction unit 506 to execute cooling of the SoCBox400. For example, the higher the temperature of the SoCBox400, the more cooling units the cooling execution unit 508 uses to execute cooling of the SoCBox400. As a specific example, when it is predicted that the temperature of the SoCBox400 will exceed a first threshold, the cooling execution unit 508 uses one of the plurality of cooling units to start cooling. Even so, when the temperature of the SoCBox400 still rises and is predicted to exceed a second threshold, the number of cooling units used is increased.

[0238] The cooling execution unit 508 can use a more powerful cooling unit to cool the SoCBox400 when the temperature of the SoCBox400 is higher. For example, when it is predicted that the temperature of the SoCBox400 will exceed the first threshold, the cooling execution unit 508 starts cooling using the air cooling unit. When the temperature of the SoCBox400 still rises despite this and is predicted to exceed the second threshold, the cooling execution unit 508 starts cooling using the water cooling unit. When the temperature of the SoCBox400 still rises despite this and is predicted to exceed the third threshold, the cooling execution unit 508 starts cooling using the liquid nitrogen cooling unit.

[0239] The SoCBox 400 may have a plurality of processing chips, and the plurality of processing chips may be respectively arranged at different positions of the SoCBox 400. Each of the plurality of cooling units may be arranged at a position corresponding to each of the plurality of processing chips.

[0240] For example, when the number of processing chips used varies according to the control conditions of autonomous driving, cooling is performed using a cooling unit corresponding to the processing chips used, thereby enabling efficient cooling.

[0241] Fig.10 An example of the SoCBox 400 and the cooling unit 600 is schematically shown. Fig.10The cooling unit 600 is illustrated as being composed of one cooling unit. When the cooling execution device 500 predicts that the SoCBox 400 starts to generate heat or that the temperature of the SoCBox 400 exceeds a preset threshold, the cooling unit 600 starts cooling, thereby cooling the entire SoCBox 400.

[0242] Fig.11 An example of the SoCBox 400 and the cooling unit 600 is schematically shown. Fig.11 The example shows a case where the cooling unit 600 is composed of a plurality of cooling units for cooling a plurality of parts of the SoCBox 400. The cooling execution device 500 predicts the temperature change of each of the plurality of parts of the SoCBox 400, and in response to the prediction that any part starts to generate heat or the temperature of any part will exceed a preset threshold, performs cooling using only the cooling unit corresponding to the part, thereby achieving efficient cooling.

[0243] Furthermore, the cooling execution device 500 of the system 10 starts cooling of the SoC Box 400 based on the temperature change of the SoC Box 400 predicted based on the position information of the vehicle 300 and the temperature information of the SoC Box 400 , thereby realizing efficient cooling according to the position of the vehicle 300 .

[0244] In addition, the cooling execution device 500 of the system 10 starts cooling the SoCBox400 based on the temperature change of the SoCBox400 predicted based on the position information of the vehicle 300 and the temperature information of the SoCBox400, and a mapping diagram representing the relationship between the position of the vehicle 300 and the temperature of the SoCBox400, thereby achieving efficient cooling corresponding to the position of the vehicle 300.

[0245] In addition, the cooling execution device 500 of the system 10 starts cooling the SoCBox400 based on the predicted temperature change of the SoCBox400 based on the position information of the vehicle 300, the information related to the type of the vehicle 300, the temperature information of the SoCBox400, and the mapping diagram representing the relationship between the position of the vehicle 300 and the temperature of the SoCBox400, thereby realizing efficient cooling corresponding to the position and model of the vehicle 300.

[0246] Furthermore, the cooling execution device 500 of the system 10 starts cooling the SoC Box 400 a given time before the SoC Box 400 starts to generate heat based on the prediction result of the temperature change of the SoC Box 400, thereby reducing the possibility of overheating and enabling efficient cooling.

[0247] In addition, the cooling execution device 500 of the system 10 uses the prediction result of the temperature change of the SoCBox 400 and performs cooling using multiple cooling units according to the temperature change of the SoCBox 400, thereby realizing efficient cooling corresponding to the temperature change.

[0248] In addition, the cooling execution device 500 of the system 10 uses the predicted results of the temperature change of the SoCBox400, and cools the SoCBox400 using a water cooling unit and a liquid nitrogen cooling unit in accordance with the temperature change of the SoCBox400, thereby being able to achieve efficient cooling using a cooling unit corresponding to the temperature change.

[0249] Fig.12 An example of the SoCBox 400 and the cooling unit 600 is schematically shown. Fig.12 The case where the cooling unit 600 is composed of two types of cooling units is illustrated. The cooling execution device 500 predicts the temperature change of each of the multiple parts of the SoCBox400, and in response to the prediction that any part begins to heat up or the temperature of any part will exceed a preset threshold, performs cooling using only the cooling unit corresponding to the part, thereby achieving efficient cooling. In addition, the cooling execution device 500 increases the cooling units used as the temperature of the SoCBox400 increases, that is, in this example, first, one of the two cooling units is used for cooling, and when the temperature of the SoCBox400 further increases, the other cooling unit is also used for cooling, thereby making the energy used for cooling more efficient.

[0250] <Third Embodiment>

[0251] When the SoC (System on Chip) for autonomous driving performs advanced computing, heat generation becomes a technical problem. Therefore, in this embodiment, a technology called Synchronized Burst Chilling is provided that flexibly applies AI (Artificial Intelligence) to optimize the rapid cooling of the SoCBox.

[0252] The SoCBox will immediately become hot, making it difficult to perform advanced calculations in the vehicle (a technical issue for fully autonomous driving). For example, there is a situation where even if the vehicle is in the same place, the heat generated by the SoCBox will change due to changes in the environment such as temperature, weather, and traffic conditions, causing overheating. This is because changes in traffic conditions, weather, etc. will increase the amount of detection by various sensors and complicate autonomous driving control. In these cases, AI predicts the heat dissipation of the SoCBox and performs cooling simultaneously with the heat dissipation, thereby preventing the SoCBox from becoming hot and making advanced calculations in the vehicle possible. It is expected that it can be used not only for cooling the SoCBox, but also for cooling the battery, which may become a solution to the high temperature caused by rapid charging.

[0253] Fig.13 An example of the system 10 is schematically shown. The system 10 includes a management server 100 . The system 10 includes a SoCBox 400 . The system 10 includes a cooling execution device 500 . The system 10 includes a cooling unit 600 .

[0254] SoCBox400, cooling execution device 500 and cooling unit 600 are mounted on the vehicle. SoCBox400 is a control device that uses the sensor values ​​of multiple sensors mounted on the vehicle to control the automatic driving of the vehicle. The automatic driving control of the vehicle consumes a very high processing load, so there is a situation where SoCBox400 becomes very hot. If the SoCBox400 is too hot, it is possible that the operation of SoCBox400 cannot be performed normally, or it may have a bad effect on the vehicle.

[0255] The system 10 involved in this embodiment predicts the temperature change of SoCBox400 based on the position information of the vehicle 300 having the cooling execution device 500, the temperature information of the SoCBox400 possessed by the vehicle 300, the information related to the traffic conditions, and the weather information, and starts cooling the SoCBox400 based on the temperature change. For example, the system 10 predicts heating in advance and immediately starts cooling the SoCBox400 based on the situation that the heating of the SoCBox400 is likely to occur due to the increase in processing volume such as traffic jams and rainy days. By starting cooling earlier than the start of heating or at the same time as the start of heating, it is possible to reliably prevent the SoCBox400 from becoming high temperature. In addition, compared with the case where the SoCBox400 is always cooled, the energy required for cooling can be reduced.

[0256] The system 10 can predict the temperature change of the SoCBox 400 through AI. The learning of the temperature change of the SoCBox 400 can be performed using data collected by the vehicle 200. For example, the management server 100 collects data from the vehicle 200 to perform learning. The subject that performs learning is not limited to the management server 100, and it can also be other devices.

[0257] The vehicle 200 is equipped with a SoCBox 400 and a temperature sensor 40 for measuring the temperature of the SoCBox 400. The SoCBox 400 controls the automatic driving of the vehicle 200 using the sensor values ​​of the multiple sensors mounted on the vehicle 200 and the external information received from the various servers 30. The server 30 may be an example of an external device. As examples of the various servers 30, a server that provides traffic information, a server that provides weather information, etc. may be cited. The SoCBox 400 sends the sensor values, external information, etc. used in the control of the automatic driving, and the temperature change of the SoCBox 400 during the control to the management server 100.

[0258] The management server 100 performs learning using information received from one or more SoCBox400. The management server 100 generates a learning model that uses the information obtained by SoCBox400 as input and the temperature change of SoCBox400 as output by performing machine learning using information such as sensor values ​​obtained by SoCBox400, external information, and the temperature change of SoCBox400 when SoCBox400 obtains such information as learning data.

[0259] The vehicle 300 is a vehicle having a cooling function according to the present embodiment. The vehicle 300 is equipped with a SoCBox 400, a cooling execution device 500, and a cooling unit 600. The cooling execution device 500 can receive and store a learning model generated by the management server 100 from the management server 100. In addition, the cooling execution device 500 can receive and store information related to traffic conditions (traffic information) from a server providing traffic information. In addition, the information related to traffic conditions includes information such as traffic congestion conditions of vehicles, the number of pedestrians, the presence or absence of accidents, the presence or absence of construction, and the presence or absence of events.

[0260] In addition, the cooling execution device 500 may receive weather information from a server that provides weather information and store it. In addition, the weather information includes information such as weather, temperature, humidity, and wind speed. In addition, the cooling execution device 500 may receive information related to the type of vehicle 300 from the server 30 and store it. In addition, the information related to the type of vehicle 300 may include information such as the model of the vehicle 300, information related to the model and the heat generation of the SoCBox 400, and parameters of each model.

[0261] The cooling execution device 500 obtains sensor values ​​of multiple sensors mounted on the vehicle 300 obtained by the SoCBox400, and external information received from multiple servers 30, or obtains it from the SoCBox400, and inputs the obtained information into the learning model, thereby predicting the temperature change of the SoCBox400. In addition, the cooling execution device 500 can perform the same processing as the prediction processing performed by the management server 100 described later.

[0262] When it is predicted that the SoC Box 400 starts to generate heat and the temperature of the SoC Box 400 will become higher than a preset threshold, the cooling execution device 500 starts cooling the SoC Box 400 using the cooling unit 600 .

[0263] SoCBox400, cooling execution device 500, management server 100, and server 30 can communicate via network 20. Network 20 can include a vehicle network. Network 20 can include the Internet. Network 20 can include a LAN (Local Area Network). Network 20 can include a mobile communication network. The mobile communication network can comply with any of the communication methods after 5G (5th Generation) communication method, LTE (Long Term Evolution) communication method, 3G (3rd Generation) communication method, and 6G (6th Generation) communication method.

[0264] Fig.14 2 is an explanatory diagram for explaining the learning stage in the system 10. Here, as the sensor 210 mounted on the vehicle 200, a camera 211, a LiDAR (Light Detection And Ranging) 212, a millimeter wave sensor 213, an ultrasonic sensor 214, an IMU sensor 215, a GNSS (Global Navigation Satellite System) sensor 216, and a GPS (Global Positioning System) sensor 217 are exemplified. The vehicle 200 does not necessarily have to be equipped with all of these sensors, and may not have some of them, or may have sensors other than these sensors.

[0265] The SoCBox 400 acquires sensor information from each sensor included in the sensor 210. In addition, the SoCBox 400 can perform communication via the network 20, and the SoCBox 400 receives external information from a plurality of servers 30 respectively via the network 20. Then, the SoCBox 400 performs automatic driving control of the vehicle 200 using the acquired information.

[0266] The temperature sensor 40 measures the temperature change of the SoCBox 400. The SoCBox 400 transmits the sensor information received from the sensor 210, the external information received from the server 30, and the temperature change measured by the temperature sensor 40 when the automatic driving control is performed by acquiring these information to the management server 100.

[0267] The management server 100 includes an information acquisition unit 102 , a model generation unit 104 , a model provision unit 106 , a creation unit 108 , and a prediction unit 110 .

[0268] The information acquisition unit 102 acquires various information. For example, the information acquisition unit 102 acquires the position information of the vehicle 300 having the cooling execution device 500, the temperature information of the SoCBox 400 possessed by the vehicle 300, the information related to the traffic conditions, and the weather information. Here, the position information may also be information acquired from a GNSS sensor or a GPS sensor. In addition, the position information may include information related to the driving route and destination of the vehicle 300.

[0269] For example, the information acquisition unit 102 acquires the position information of the vehicle 300 having the cooling execution device 500, the temperature information of the SoCBox 400 possessed by the vehicle 300, the information related to the traffic condition, and the weather information from the cooling execution device 500. In addition, for example, the information acquisition unit 102 acquires the weather information from the server providing the weather information. In addition, for example, the information acquisition unit 102 acquires the information related to the traffic condition from the server providing the traffic information.

[0270] In addition, the information acquisition unit 102 acquires information related to the type of the vehicle 300 from the cooling execution device 500. In addition, the information related to the type of the vehicle may include information such as the model of the vehicle 300, information related to the model and the heat generation of the SoCBox400, and parameters of each model. In addition, the information acquisition unit 102 can acquire various information based on a given timing or cycle. The management server 100 can receive information sent by the SoCBox400.

[0271] The model generation unit 104 generates a learning model by performing machine learning using the information acquired by the information acquisition unit 102. The model generation unit 104 generates a learning model that uses the information acquired by SoCBox400 as input and the temperature change of SoCBox400 as output by performing machine learning using the information acquired by SoCBox400 and the temperature change of SoCBox400 when SoCBox400 acquires the information as learning data.

[0272] The model providing unit 106 provides the learning model generated by the model generating unit 104. The model providing unit 106 may transmit the learning model to the cooling execution device 500 mounted on the vehicle 300.

[0273] The creation unit 108 creates a map representing the relationship between the position of the vehicle 300, the information related to the traffic conditions, the weather information, and the temperature of the SoCBox 400, based on the position information of the vehicle 300 having the cooling execution device 500, the temperature information of the SoCBox 400 of the vehicle 300, the information related to the traffic conditions, and the weather information acquired by the information acquisition unit 102. For example, the creation unit 108 creates a map representing the relationship between the position of the vehicle 300 and the temperature of the SoCBox 400 for each of the information related to the traffic conditions and the weather information, based on the position information of the vehicle 300 having the cooling execution device 500, the temperature information of the SoCBox 400 of the vehicle 300, the information related to the traffic conditions, and the weather information acquired by the information acquisition unit 102.

[0274] In addition, the creation unit 108 creates a mapping diagram representing the relationship between the position of the vehicle 300 and the temperature of the SoCBox400 for each type of vehicle 300 based on the position information of the vehicle 300 having the cooling execution device 500, the temperature information of the SoCBox400 possessed by the vehicle 300, and the information related to the type of the vehicle 300 acquired by the information acquisition unit 102.

[0275] The prediction unit 110 predicts the temperature change of the SoCBox 400 based on the position information of the vehicle 300 having the cooling execution device 500, the temperature information of the SoCBox 400 of the vehicle 300, the information related to the traffic conditions, and the weather information acquired by the information acquisition unit 102. At this time, the prediction unit 110 may also use the position information of the vehicle 300 having the cooling execution device 500, the temperature information of the SoCBox 400 of the vehicle 300, the information related to the traffic conditions, and the weather information acquired by the information acquisition unit 102 in the past to perform the prediction.

[0276] For example, the prediction unit 110 uses the accumulated position information of the vehicle 300 having the cooling execution device 500, the temperature information of the SoCBox400 possessed by the vehicle 300, the information related to the traffic conditions, and the weather information to predict the temperature change of the SoCBox400 based on the position information of the vehicle 300 having the cooling execution device 500, the temperature information of the SoCBox400 possessed by the vehicle 300, the information related to the traffic conditions, and the weather information acquired by the information acquisition unit 102.

[0277] For example, based on the data obtained in the past that the heat value increases by about 20% in rainy days compared to sunny days, the prediction unit 110 predicts the temperature change based on the basis that the heat value increases by about 20% when the weather information is rain. Also, based on the data obtained in the past that the heat value increases by about 30% in traffic jams compared to normal times, the prediction unit 110 predicts the temperature change based on the basis that the heat value increases by about 30% when the traffic condition is traffic jams, for example.

[0278] In addition, the prediction unit 110 predicts the temperature change of SoCBox400 based on the position information of the vehicle 300 having the cooling execution device 500 acquired by the information acquisition unit 102, the temperature information of the SoCBox400 possessed by the vehicle 300, the information related to the traffic conditions, the weather information, and the mapping diagram created by the creation unit 108 that represents the relationship between the position of the vehicle 300, the information related to the traffic conditions, the weather information and the temperature of the SoCBox400.

[0279] In addition, the prediction unit 110 predicts the temperature change of SoCBox400 based on the position information of the vehicle 300 having the cooling execution device 500 acquired by the information acquisition unit 102, the temperature information of the SoCBox400 possessed by the vehicle 300, the information related to the traffic condition, the weather information, and the mapping map created by the creation unit 108, which characterizes the relationship between the position of the vehicle 300 and the temperature of the SoCBox400 for each of the information related to the traffic condition and the weather information.

[0280] In addition, the prediction unit 110 predicts the temperature change of SoCBox400 based on the position information of the vehicle 300 having the cooling execution device 500 acquired by the information acquisition unit 102, the temperature information of the SoCBox400 possessed by the vehicle 300, and the mapping diagram created by the creation unit 108 that represents the relationship between the position of the vehicle 300 and the temperature of the SoCBox400 in the type of the vehicle 300.

[0281] The temperature change of SoCBox400 predicted by the prediction unit 110 includes the temperature change of SoCBox400 of the vehicle 300 over time. For example, the prediction unit 110 may predict the temperature change of SoCBox400 of the vehicle 300 over time on the travel route from the current position of the vehicle 300 to the destination.

[0282] The system 10 may also be configured to predict the temperature change of each of the multiple parts of the SoCBox400. In this case, the vehicle 200 may be provided with multiple temperature sensors 40 that measure the temperature change of each of the multiple parts of the SoCBox400. The SoCBox400 may send the sensor information received from the sensor 210, the external information received from the server 30, and the temperature change measured by the multiple temperature sensors 40 when the automatic driving control is performed after obtaining the information to the management server 100. The model generation unit 104 performs machine learning by using the information obtained by the SoCBox400 and the temperature change of each of the multiple parts of the SoCBox400 when the SoCBox400 obtains the information as learning data, thereby generating a learning model that uses the information obtained by the SoCBox400 as input and the temperature change of each of the multiple parts of the SoCBox400 as output.

[0283] Fig.15 3 is an explanatory diagram for explaining the cooling execution stage in the system 10. Here, as sensors 310 mounted on the vehicle 300, a camera 311, a LiDAR 312, a millimeter wave sensor 313, an ultrasonic sensor 314, an IMU sensor 315, a GNSS sensor 316, and a GPS sensor 317 are exemplified. The vehicle 300 does not necessarily have to be equipped with all of these sensors, and may not have some of them, or may have sensors other than these sensors.

[0284] The cooling execution device 500 includes a model storage unit 502, an information acquisition unit 504, a prediction unit 506, and a cooling execution unit 508. The model storage unit 502 stores the learning model received from the management server 100. The information acquisition unit 504 acquires the information acquired by the SoCBox 400.

[0285] The information acquisition unit 504 acquires the sensor information acquired by the SoCBox400 from the sensor 310 from the sensor 310 or the SoCBox400. For example, the information acquisition unit 504 may receive the sensor information acquired by the SoCBox400 from the sensor 310 from the SoCBox400. The information acquisition unit 504 may also receive the same sensor information from the sensor 310 as the sensor information acquired by the SoCBox400 from the sensor 310. In this case, each of the sensors 310 may send the sensor information to the SoCBox400 and each of the cooling execution devices 500.

[0286] The information acquisition unit 504 acquires the external information acquired by the SoCBox400 from the server 30 from the server 30 or the SoCBox400. The information acquisition unit 504 may receive the external information received by the SoCBox400 from the server 30 from the SoCBox400. The information acquisition unit 504 may also receive the same external information from the server 30 as the external information received by the SoCBox400 from the server 30. In this case, the server 30 may send the external information to each of the SoCBox400 and the cooling execution device 500.

[0287] The information acquisition unit 504 acquires, from the management server 100 , the prediction result of the temperature change of the SoCBox 400 predicted by the prediction unit 110 of the management server 100 .

[0288] The prediction unit 506 predicts the temperature change of the SoCBox400. The prediction unit 506 can predict the temperature change of the SoCBox400 through AI. For example, the prediction unit 506 predicts the temperature change of the SoCBox400 by inputting the information acquired by the information acquisition unit 504 into the learning model stored in the model storage unit 502. In addition, the prediction unit 506 can perform the same prediction processing as the prediction processing performed by the prediction unit 110 of the management server 100.

[0289] The cooling execution unit 508 starts cooling the SoCBox400 based on the prediction result of the temperature change of the SoCBox400 acquired by the information acquisition unit 504. For example, the cooling execution unit 508 uses the prediction result of the temperature change of the SoCBox400 acquired by the information acquisition unit 504 to start cooling the SoCBox400 from a given time before the SoCBox400 starts to generate heat. In addition, for example, the cooling execution unit 508 uses the prediction result of the temperature change of the SoCBox400 acquired by the information acquisition unit 504 to start cooling the SoCBox400 from a given time before the timing when the SoCBox400 is predicted to become a given temperature or above. In addition, for example, the cooling execution unit 508 uses the prediction result of the temperature change of the SoCBox400 acquired by the information acquisition unit 504 to start cooling the SoCBox400 when the location where the SoCBox400 is predicted to start generating heat and the current position of the vehicle 300 become within a given distance.

[0290] The cooling execution unit 508 starts cooling the SoCBox 400 based on the temperature change of the SoCBox 400 predicted by the prediction unit 506. For example, the cooling execution unit 508 starts cooling the SoCBox 400 in response to the prediction unit 506 predicting that the SoCBox 400 starts to heat up. For example, the cooling execution unit 508 starts cooling the SoCBox 400 in response to the prediction unit 506 predicting that the temperature of the SoCBox 400 will become higher than a preset threshold.

[0291] The cooling execution unit 508 uses the predicted result of the temperature change of the SoCBox400 obtained by the information acquisition unit 504, and uses one or more cooling units selected from a plurality of cooling units to start cooling the SoCBox400 according to the temperature change of the SoCBox400. The cooling execution unit 508 can use the cooling unit 600 to execute the cooling of the SoCBox400. The cooling unit 600 can use an air cooling unit to cool the SoCBox400. The cooling unit 600 can use a water cooling unit to cool the SoCBox400. The cooling unit 600 can use a liquid nitrogen cooling unit to cool the SoCBox400.

[0292] The cooling unit 600 may also include multiple cooling units. For example, the cooling unit 600 includes multiple air cooling units. For example, the cooling unit 600 includes multiple water cooling units. For example, the cooling unit 600 includes multiple liquid nitrogen cooling units. The cooling unit 600 may also include multiple of one or more air cooling units, one or more water cooling units, and one or more liquid nitrogen cooling units.

[0293] The plurality of cooling units may be configured to cool different parts of the SoCBox 400. The prediction unit 506 may use the information acquired by the information acquisition unit 504 to predict the temperature change of each of the plurality of parts of the SoCBox 400. The cooling execution unit 508 may start cooling the SoCBox 400 using a cooling unit selected from the plurality of cooling units that cool the plurality of parts of the SoCBox 400, based on the prediction result of the prediction unit 506.

[0294] The cooling execution unit 508 can use a cooling unit corresponding to the temperature of the SoCBox400 predicted by the prediction unit 506 to execute cooling of the SoCBox400. For example, the higher the temperature of the SoCBox400, the more cooling units the cooling execution unit 508 uses to execute cooling of the SoCBox400. As a specific example, when it is predicted that the temperature of the SoCBox400 will exceed a first threshold, the cooling execution unit 508 uses one of the plurality of cooling units to start cooling. Even so, when the temperature of the SoCBox400 still rises and is predicted to exceed a second threshold, the number of cooling units used is increased.

[0295] The cooling execution unit 508 can use a more powerful cooling unit to cool the SoCBox400 when the temperature of the SoCBox400 is higher. For example, when it is predicted that the temperature of the SoCBox400 will exceed the first threshold, the cooling execution unit 508 starts cooling using the air cooling unit. When the temperature of the SoCBox400 still rises despite this and is predicted to exceed the second threshold, the cooling execution unit 508 starts cooling using the water cooling unit. When the temperature of the SoCBox400 still rises despite this and is predicted to exceed the third threshold, the cooling execution unit 508 starts cooling using the liquid nitrogen cooling unit.

[0296] The SoCBox 400 may have a plurality of processing chips, and the plurality of processing chips may be respectively arranged at different positions of the SoCBox 400. Each of the plurality of cooling units may be arranged at a position corresponding to each of the plurality of processing chips.

[0297] For example, when the number of processing chips used varies according to the control conditions of autonomous driving, cooling is performed using a cooling unit corresponding to the processing chips used, thereby enabling efficient cooling.

[0298] Fig.16 An example of the SoCBox 400 and the cooling unit 600 is schematically shown. Fig.16The cooling unit 600 is configured by one cooling unit. When the cooling execution device 500 predicts that the SoCBox 400 starts to generate heat or that the temperature of the SoCBox 400 exceeds a preset threshold, the cooling execution device 500 starts cooling using the cooling unit 600 to cool the entire SoCBox 400.

[0299] Fig.17 An example of the SoCBox 400 and the cooling unit 600 is schematically shown. Fig.17 The example shows a case where the cooling unit 600 is composed of a plurality of cooling units for cooling a plurality of parts of the SoCBox 400. The cooling execution device 500 predicts the temperature change of each of the plurality of parts of the SoCBox 400, and in response to the prediction that any part starts to generate heat or the temperature of any part will exceed a preset threshold, performs cooling using only the cooling unit corresponding to the part, thereby achieving efficient cooling.

[0300] In addition, the cooling execution device 500 of the system 10 starts cooling the SoCBox400 according to the temperature change of the SoCBox400 predicted based on the position information of the vehicle 300, the temperature information of the SoCBox400, the information related to the traffic conditions, and the weather information, thereby being able to achieve efficient cooling in response to the temperature change of the SoCBox400 when the detection amount of various sensors increases due to changes in the environment and when the autonomous driving control becomes complicated.

[0301] In addition, the cooling execution device 500 of the system 10 starts cooling the SoCBox400 based on the predicted temperature change of the SoCBox400 based on the position information of the vehicle 300, the temperature information of the SoCBox400, the information related to the traffic conditions, the weather information, and the mapping diagram representing the relationship between the position information of the vehicle 300, the information related to the traffic conditions, the weather information and the temperature of the SoCBox400, thereby realizing efficient cooling corresponding to the position of the vehicle 300 and the dynamic environment.

[0302] In addition, the cooling execution device 500 of the system 10 creates a mapping diagram representing the relationship between the position information of the vehicle 300 and the temperature of the SoCBox400 according to each piece of information related to traffic conditions and weather information to predict the temperature change of the SoCBox400, and starts cooling the SoCBox400, thereby achieving efficient cooling corresponding to the position of the vehicle 300 in each dynamic environment.

[0303] In addition, the cooling execution device 500 of the system 10 starts cooling the SoCBox400 based on the predicted temperature change of the SoCBox400 based on the position information of the vehicle 300, the temperature information of the SoCBox400, the information related to the traffic conditions, the weather information, the model of the vehicle 300, and the mapping diagram representing the relationship between the position information of the vehicle 300, the information related to the traffic conditions, the weather information and the temperature of the SoCBox400, thereby realizing efficient cooling corresponding to the model, position and dynamic environment of the vehicle 300.

[0304] Furthermore, the cooling execution device 500 of the system 10 starts cooling the SoC Box 400 a given time before the SoC Box 400 starts to generate heat based on the prediction result of the temperature change of the SoC Box 400, thereby reducing the possibility of overheating and enabling efficient cooling.

[0305] In addition, the cooling execution device 500 of the system 10 uses the prediction result of the temperature change of the SoCBox 400 and performs cooling using multiple cooling units according to the temperature change of the SoCBox 400, thereby realizing efficient cooling corresponding to the temperature change.

[0306] In addition, the cooling execution device 500 of the system 10 uses the predicted results of the temperature change of the SoCBox400, and cools the SoCBox400 using a water cooling unit and a liquid nitrogen cooling unit in accordance with the temperature change of the SoCBox400, thereby being able to achieve efficient cooling using a cooling unit corresponding to the temperature change.

[0307] Fig.18 An example of the SoCBox 400 and the cooling unit 600 is schematically shown. Fig.18 The case where the cooling unit 600 is composed of two types of cooling units is illustrated. The cooling execution device 500 predicts the temperature change of each of the multiple parts of the SoCBox400, and in response to the prediction that any part begins to heat up or the temperature of any part will exceed a preset threshold, performs cooling using only the cooling unit corresponding to the part, thereby achieving efficient cooling. In addition, the cooling execution device 500 increases the cooling units used as the temperature of the SoCBox400 increases, that is, in this example, first, one of the two cooling units is used for cooling, and when the temperature of the SoCBox400 further increases, the other cooling unit is also used for cooling, thereby making the energy used for cooling more efficient.

[0308] <Fourth embodiment>

[0309] When SoC (System on Chip) for autonomous driving performs advanced computing, heat generation becomes a technical problem. Therefore, in this embodiment, a technology, namely Synchronized Burst Chilling, is provided that flexibly applies AI (Artificial Intelligence) to optimize the rapid cooling of a SoCBox having multiple SoCs.

[0310] The SoCBox will immediately become hot, making it difficult to perform advanced calculations in the vehicle (a technical issue for fully autonomous driving). For example, there is a situation where the SoCBox overheats due to changes in operating conditions, an increase in the amount of detection by various sensors, and the complication of autonomous driving control. The SoCBox with multiple SoCs (integrated circuits) generates heat unevenly. For example, when a certain autonomous driving control is performed, only the SoC responsible for the autonomous driving control generates heat. In these cases, AI predicts the heat dissipation of the SoCBox and its location, and cools the location where the heat dissipation occurs simultaneously with the heat dissipation, thereby preventing the SoCBox from becoming hot and making advanced calculations in the vehicle possible. Not only the SoCBox, but it is also expected to be used for cooling the battery, which may become a solution to the high temperature caused by rapid charging.

[0311] Fig.19 An example of the system 10 is schematically shown. The system 10 includes a management server 100 . The system 10 includes a SoCBox 400 . The system 10 includes a cooling execution device 500 . The system 10 includes a cooling unit 600 .

[0312] SoCBox400, cooling execution device 500 and cooling unit 600 are mounted on the vehicle. SoCBox400 has multiple SoCs and is a control device that uses the sensor values ​​of multiple sensors mounted on the vehicle to control the automatic driving of the vehicle. The automatic driving control of the vehicle consumes a very high processing load, so there is a situation where SoCBox400 becomes very hot. If the SoCBox400 is too hot, it is possible that the operation of SoCBox400 cannot be performed normally, or it may have a negative impact on the vehicle.

[0313] The system 10 involved in this embodiment predicts the temperature change of each position of SoCBox400 based on the position information of the SoC in SoCBox400, the information related to the processing of the SoC, and the information related to the driving condition of the vehicle, and starts cooling the SoCBox400 based on the temperature change of each position. For example, the system 10 predicts the heat generated and its position in various control processes such as straight driving, right and left turns, stopping, and condition judgment using various sensors, and immediately starts cooling the SoCBox400. By starting cooling earlier than the start of heating or at the same time as the start of heating, it is possible to reliably prevent the SoCBox400 from becoming high temperature. In addition, compared with the case where the SoCBox400 is always cooled, the energy required for cooling can be reduced.

[0314] The system 10 can predict the temperature change of each position of the SoCBox400 through AI. The learning of the temperature change of each position of the SoCBox400 can be performed using the data collected by the vehicle 200. For example, the management server 100 collects data from the vehicle 200 to perform learning. The subject that performs learning is not limited to the management server 100, and it can also be other devices.

[0315] The vehicle 200 is equipped with a SoCBox400 and a temperature sensor 40 for measuring the temperature at each position of the SoCBox400. The SoCBox400 uses the sensor values ​​of the multiple sensors mounted on the vehicle 200 and the external information received from the various servers 30 to control the automatic driving of the vehicle 200. In addition, the SoCBox400 stores the position information of the SoC contained in the SoCBox400, the information related to the processing of each SoC, and the information related to the operation status of the vehicle 300. Here, the information related to the processing of each SoC refers to the information of the processing and control performed by each SoC, including the information of the acceleration, deceleration, right and left turns, stop, driving vehicle detection, pedestrian detection, obstacle detection, detection performed by various sensors, various calculations, etc. controlled by each SoC. In addition, the information related to the operation status of the vehicle 300 refers to the information of the operation status and operation status of the vehicle 300, including the information of the acceleration, deceleration, right and left turns, stop, driving vehicle detection, pedestrian detection, obstacle detection, detection performed by various sensors, various calculations, etc. In addition, the above-mentioned processing content and operating conditions are only examples, and also include processing and operating conditions other than the recorded content. The server 30 can be an example of an external device. As examples of various servers 30, a server that provides traffic information, a server that provides weather information, etc. can be listed. SoCBox400 sends sensor values, external information, etc. used in the control of autonomous driving, and temperature changes at various locations of SoCBox400 during control to the management server 100.

[0316] The management server 100 performs learning using information received from one or more SoCBox400. The management server 100 generates a learning model that uses the information obtained by SoCBox400 as input and the temperature change at each position of SoCBox400 as output by performing machine learning using information such as sensor values ​​obtained by SoCBox400, external information, and the temperature change at each position of SoCBox400 when SoCBox400 obtains the information as learning data.

[0317] The vehicle 300 is a vehicle having a cooling function according to the present embodiment. The vehicle 300 is equipped with a SoCBox 400, a cooling execution device 500, and a cooling unit 600. The cooling execution device 500 can receive a learning model generated by the management server 100 from the management server 100 and store the learning model.

[0318] In addition, the cooling execution device 500 may store the position information of each SoC in the SoC Box 400 acquired from the SoC Box 400 , the information related to the processing of each SoC, and the information related to the operating status of the vehicle 300 .

[0319] The cooling execution device 500 obtains sensor values ​​of multiple sensors mounted on the vehicle 300 obtained by the SoCBox400, external information received from multiple servers 30, or from the SoCBox400, and inputs the obtained information into the learning model, thereby predicting the temperature change at each position of the SoCBox400. In addition, the cooling execution device 500 can perform the same processing as the prediction processing performed by the management server 100 described later.

[0320] When the cooling execution device 500 predicts that the SoC Box 400 starts to generate heat or predicts that the temperature of each position of the SoC Box 400 will become higher than a preset threshold, the cooling execution device 500 starts cooling the SoC Box 400 using the cooling unit 600 .

[0321] SoCBox400, cooling execution device 500, management server 100, and server 30 can communicate via network 20. Network 20 can include a vehicle network. Network 20 can include the Internet. Network 20 can include a LAN (Local Area Network). Network 20 can include a mobile communication network. The mobile communication network can comply with any of the communication methods after 5G (5th Generation) communication method, LTE (Long Term Evolution) communication method, 3G (3rd Generation) communication method, and 6G (6th Generation) communication method.

[0322] Fig. 20 2 is an explanatory diagram for explaining the learning stage in the system 10. Here, as the sensor 210 mounted on the vehicle 200, a camera 211, a LiDAR (Light Detection And Ranging) 212, a millimeter wave sensor 213, an ultrasonic sensor 214, an IMU sensor 215, a GNSS (Global Navigation Satellite System) sensor 216, and a GPS (Global Positioning System) sensor 217 are exemplified. The vehicle 200 does not necessarily have to be equipped with all of these sensors, and may not have some of them, or may have sensors other than these sensors.

[0323] SoCBox400 acquires sensor information from each sensor included in the sensor 210. In addition, SoCBox400 can perform communication via the network 20, and SoCBox400 receives external information from multiple servers 30 via the network 20. Then, SoCBox400 uses the acquired information to perform automatic driving control of the vehicle 200. In addition, SoCBox400 stores the position information of each SoC possessed by SoCBox400.

[0324] The temperature sensor 40 measures temperature changes at various locations of the SoCBox 400. The SoCBox 400 sends sensor information received from the sensor 210, external information received from the server 30, and temperature changes at various locations measured by the temperature sensor 40 when acquiring these information to perform autonomous driving control to the management server 100.

[0325] The management server 100 includes an information acquisition unit 102 , a model generation unit 104 , a model provision unit 106 , and a prediction unit 110 .

[0326] The information acquisition unit 102 acquires various information. For example, the information acquisition unit 102 acquires the position information of each SoC in the SoCBox400, the information related to the processing of each SoC, and the information related to the operating status of the vehicle 300. For example, the information acquisition unit 102 acquires the position information of each SoC in the SoCBox400, the information related to the processing of each SoC, and the information related to the operating status of the vehicle 300 from the cooling execution device 500. For example, the information acquisition unit 102 can also use information such as the manufacturer and model of the SoCBox400 to determine the position of each SoC and thereby acquire the position information of each SoC. In addition, the information acquisition unit 102 can acquire various information based on a given timing or cycle. The management server 100 can receive information sent by the SoCBox400.

[0327] The model generation unit 104 generates a learning model by performing machine learning using the information acquired by the information acquisition unit 102. The model generation unit 104 generates a learning model that uses the information acquired by SoCBox400 as input and the temperature change at each position of SoCBox400 as output by performing machine learning using the information acquired by SoCBox400 and the temperature change at each position of SoCBox400 when SoCBox400 acquires the information as learning data.

[0328] The model providing unit 106 provides the learning model generated by the model generating unit 104. The model providing unit 106 may transmit the learning model to the cooling execution device 500 mounted on the vehicle 300.

[0329] The prediction unit 110 predicts temperature changes at various locations of the SoCBox400 based on the position information of each SoC in the SoCBox400 acquired by the information acquisition unit 102, information related to the processing of each SoC, and information related to the operating status of the vehicle 300. For example, the prediction unit 110 predicts temperature changes at various locations of the SoCBox400 using information related to the operating status acquired by the information acquisition unit 102, such as information on driving while performing obstacle detection and pedestrian detection, information related to the processing of each SoC, information on the SoC that controls obstacle detection, pedestrian detection, and driving, and position information of the SoC in the SoCBox400.

[0330] The prediction unit 110 uses the position information of each SoC in the SoCBox 400 , information related to the processing of each SoC, and information related to the operating status of the vehicle 300 acquired by the information acquisition unit 102 to predict the temperature change over time at each position of the SoCBox 400 .

[0331] For example, the prediction unit 110 uses information related to the operating condition, such as information on driving while performing obstacle detection and pedestrian detection, acquired by the information acquisition unit 102, information related to the processing of each SoC, such as information on the SoC that controls obstacle detection, pedestrian detection, and driving, and position information of the SoC in SoCBox400, to predict the temperature changes at various locations of SoCBox400 when the operating condition continues.

[0332] The system 10 may also be configured to predict the temperature change of each of the multiple parts of the SoCBox400. In this case, the vehicle 200 may be provided with multiple temperature sensors 40 that measure the temperature change of each of the multiple parts of the SoCBox400. The SoCBox400 may send the sensor information received from the sensor 210, the external information received from the server 30, and the temperature change measured by the multiple temperature sensors 40 when the automatic driving control is performed after obtaining the information to the management server 100. The model generation unit 104 performs machine learning by using the information obtained by the SoCBox400 and the temperature change of each of the multiple parts of the SoCBox400 when the SoCBox400 obtains the information as learning data, thereby generating a learning model that uses the information obtained by the SoCBox400 as input and the temperature change of each of the multiple parts of the SoCBox400 as output.

[0333] Fig.21 3 is an explanatory diagram for explaining the cooling execution stage in the system 10. Here, as sensors 310 mounted on the vehicle 300, a camera 311, a LiDAR 312, a millimeter wave sensor 313, an ultrasonic sensor 314, an IMU sensor 315, a GNSS sensor 316, and a GPS sensor 317 are exemplified. The vehicle 300 does not necessarily have to be equipped with all of these sensors, and may not have some of them, or may have sensors other than these sensors.

[0334] The cooling execution device 500 includes a model storage unit 502, an information acquisition unit 504, a prediction unit 506, and a cooling execution unit 508. The model storage unit 502 stores the learning model received from the management server 100. The information acquisition unit 504 acquires the information acquired by the SoCBox 400.

[0335] The information acquisition unit 504 acquires the sensor information acquired by the SoCBox400 from the sensor 310 from the sensor 310 from the sensor 310 or the SoCBox400. For example, the information acquisition unit 504 may receive the sensor information acquired by the SoCBox400 from the sensor 310 from the SoCBox400. The information acquisition unit 504 may also receive the same sensor information from the sensor 310 as the sensor information acquired by the SoCBox400 from the sensor 310. In this case, each of the sensors 310 may send the sensor information to the SoCBox400 and each of the cooling execution devices 500. In addition, for example, the information acquisition unit 504 acquires the information of the position of each SoC in the SoCBox400 from the SoCBox400.

[0336] The information acquisition unit 504 acquires the external information acquired by the SoCBox400 from the server 30 from the server 30 or the SoCBox400. The information acquisition unit 504 may receive the external information received by the SoCBox400 from the server 30 from the SoCBox400. The information acquisition unit 504 may also receive the same external information from the server 30 as the external information received by the SoCBox400 from the server 30. In this case, the server 30 may send the external information to each of the SoCBox400 and the cooling execution device 500.

[0337] The information acquisition unit 504 acquires, from the management server 100 , the prediction result of the temperature change at each position of the SoCBox 400 predicted by the prediction unit 110 of the management server 100 .

[0338] The prediction unit 506 predicts the temperature change of the SoCBox400. The prediction unit 506 can predict the temperature change of the SoCBox400 through AI. For example, the prediction unit 506 predicts the temperature change of the SoCBox400 by inputting the information acquired by the information acquisition unit 504 into the learning model stored in the model storage unit 502. In addition, the prediction unit 506 can perform the same prediction processing as the prediction processing performed by the prediction unit 110 of the management server 100.

[0339] The cooling execution unit 508 starts cooling the SoCBox400 based on the predicted results of the temperature changes at various locations of the SoCBox400 acquired by the information acquisition unit 504. For example, the cooling execution unit 508 uses the predicted results of the temperature changes at various locations of the SoCBox400 acquired by the information acquisition unit 504 to start cooling the location of the SoCBox400 where heat is generated.

[0340] Furthermore, for example, the cooling execution unit 508 uses the prediction results of the temperature change at each position of the SoC Box 400 acquired by the information acquisition unit 504 to start cooling the position where heat generation of the SoC Box 400 occurs from a predetermined time before the SoC Box 400 starts to generate heat.

[0341] In addition, for example, the cooling execution unit 508 uses the predicted results of the temperature changes at various locations of the SoCBox400 acquired by the information acquisition unit 504 to start cooling the locations where heat generation occurs in the SoCBox400 from a given time before the timing when the temperature at various locations of the SoCBox400 is predicted to become higher than a given temperature.

[0342] The cooling execution unit 508 starts cooling the SoCBox 400 based on the temperature change of the SoCBox 400 predicted by the prediction unit 506. For example, the cooling execution unit 508 starts cooling the SoCBox 400 in response to the prediction unit 506 predicting that the SoCBox 400 starts to heat up. For example, the cooling execution unit 508 starts cooling the SoCBox 400 in response to the prediction unit 506 predicting that the temperature of the SoCBox 400 will become higher than a preset threshold.

[0343] The cooling execution unit 508 uses the predicted results of the temperature changes at various locations of the SoCBox400 acquired by the information acquisition unit 504, and uses one or more cooling units selected from a plurality of cooling units to start cooling the locations where heat generation of the SoCBox400 occurs, corresponding to the temperature changes at various locations of the SoCBox400. The cooling execution unit 508 can use the cooling unit 600 to execute cooling of the SoCBox400. The cooling unit 600 can use an air cooling unit to cool the SoCBox400. The cooling unit 600 can use a water cooling unit to cool the SoCBox400. The cooling unit 600 can use a liquid nitrogen cooling unit to cool the locations where heat generation of the SoCBox400 occurs.

[0344] The cooling unit 600 may also include multiple cooling units. For example, the cooling unit 600 includes multiple air cooling units. For example, the cooling unit 600 includes multiple water cooling units. For example, the cooling unit 600 includes multiple liquid nitrogen cooling units. The cooling unit 600 may also include multiple of one or more air cooling units, one or more water cooling units, and one or more liquid nitrogen cooling units.

[0345] The plurality of cooling units may be configured to cool different parts of the SoCBox 400. The prediction unit 506 may use the information acquired by the information acquisition unit 504 to predict the temperature change of each of the plurality of parts of the SoCBox 400. The cooling execution unit 508 may start cooling the location where heat generation of the SoCBox 400 occurs, using a cooling unit selected from the plurality of cooling units that cool the plurality of parts of the SoCBox 400, based on the prediction result of the prediction unit 506.

[0346] The cooling execution unit 508 can use a cooling unit corresponding to the temperature of the SoCBox400 predicted by the prediction unit 506 to execute cooling of the SoCBox400. For example, the higher the temperature of the SoCBox400, the more cooling units the cooling execution unit 508 uses to cool the location where the heat of the SoCBox400 occurs. As a specific example, when it is predicted that the temperature of the SoCBox400 will exceed the first threshold, the cooling execution unit 508 uses one of the multiple cooling units to start cooling. Even so, when the temperature of the SoCBox400 still rises and is predicted to exceed the second threshold, the number of cooling units used is increased.

[0347] The cooling execution unit 508 can use a more powerful cooling unit to cool the SoCBox400 when the temperature of the SoCBox400 is higher. For example, when it is predicted that the temperature of the SoCBox400 will exceed the first threshold, the cooling execution unit 508 starts cooling using the air cooling unit. When the temperature of the SoCBox400 still rises despite this and is predicted to exceed the second threshold, the cooling execution unit 508 starts cooling using the water cooling unit. When the temperature of the SoCBox400 still rises despite this and is predicted to exceed the third threshold, the cooling execution unit 508 starts cooling using the liquid nitrogen cooling unit.

[0348] The SoCBox 400 may have a plurality of processing chips, and the plurality of processing chips may be respectively arranged at different positions of the SoCBox 400. Each of the plurality of cooling units may be arranged at a position corresponding to each of the plurality of processing chips.

[0349] For example, when the number of processing chips used varies according to the control conditions of autonomous driving, cooling is performed using a cooling unit corresponding to the processing chips used, thereby enabling efficient cooling.

[0350] Fig. 22 An example of the SoCBox 400 and the cooling unit 600 is schematically shown. Fig. 22The cooling unit 600 is configured by one cooling unit. When the cooling execution device 500 predicts that the SoCBox 400 starts to generate heat or that the temperature of the SoCBox 400 exceeds a preset threshold, the cooling execution device 500 starts cooling using the cooling unit 600 to cool the entire SoCBox 400.

[0351] Fig.23 An example of the SoCBox 400 and the cooling unit 600 is schematically shown. Fig.23 The example shows a case where the cooling unit 600 is composed of a plurality of cooling units for cooling a plurality of parts of the SoCBox 400. The cooling execution device 500 predicts the temperature change of each of the plurality of parts of the SoCBox 400, and in response to the prediction that any part starts to generate heat or the temperature of any part will exceed a preset threshold, performs cooling using only the cooling unit corresponding to the part, thereby achieving efficient cooling.

[0352] The cooling execution device 500 of the system 10 starts cooling the location where heat occurs in the SoCBox400 based on the predicted results of the temperature changes at various locations of the SoCBox400 predicted according to the location information of the SoC in the SoCBox400, the information related to the control of the SoC, and the operating conditions of the vehicle 300, thereby cooling only the places that need cooling, thereby achieving efficient cooling.

[0353] In addition, the cooling execution device 500 of the system 10 uses the predicted results of the temperature changes over time at each position of the SoCBox400 predicted based on the position information of the SoC in the SoCBox400, the information related to the control of the SoC, and the operating conditions of the vehicle 300, to start cooling the position where the heat of the SoCBox400 occurs, thereby cooling only the places that need cooling at an effective time, thereby achieving efficient cooling.

[0354] In addition, the cooling execution device 500 of the system 10 uses the predicted results of the temperature changes at various positions of the SoCBox400 predicted based on the position information of the SoC in the SoCBox400, the information related to the control of the SoC, and the operating conditions of the vehicle 300, and starts cooling the position where the heat of the SoCBox400 occurs before the heat occurs, thereby preventing overheating while cooling only the places that need cooling, thereby achieving efficient cooling.

[0355] In addition, the cooling execution device 500 of the system 10 uses the prediction results of the temperature changes at various positions of the SoCBox400 and uses multiple cooling units to perform cooling according to the temperature changes at various positions of the SoCBox400, thereby achieving efficient cooling corresponding to the temperature changes.

[0356] In addition, the cooling execution device 500 of the system 10 uses the predicted results of the temperature changes at various positions of the SoCBox400, and uses a water cooling unit and a liquid nitrogen cooling unit to cool down in accordance with the temperature changes at various positions of the SoCBox400, thereby being able to achieve efficient cooling using a cooling unit corresponding to the temperature changes.

[0357] Fig.24 An example of the SoCBox 400 and the cooling unit 600 is schematically shown. Fig.24 The case where the cooling unit 600 is composed of two types of cooling units is illustrated. The cooling execution device 500 predicts the temperature change of each of the multiple parts of the SoCBox400, and in response to the prediction that any part begins to heat up or the temperature of any part will exceed a preset threshold, performs cooling using only the cooling unit corresponding to the part, thereby achieving efficient cooling. In addition, the cooling execution device 500 increases the cooling units used as the temperature of the SoCBox400 increases, that is, in this example, first, one of the two cooling units is used for cooling, and when the temperature of the SoCBox400 further increases, the other cooling unit is also used for cooling, thereby making the energy used for cooling more efficient.

[0358] <Fifth embodiment>

[0359] When the SoC (System on Chip) for autonomous driving performs advanced computing, heat generation becomes a technical problem. Therefore, in this embodiment, a technology called Synchronized Burst Chilling is provided that flexibly applies AI (Artificial Intelligence) to optimize the rapid cooling of the SoCBox.

[0360] The SoCBox will immediately become hot, making it difficult to perform advanced calculations in the vehicle (a technical issue for fully autonomous driving). AI predicts the heat dissipation of the SoCBox and performs cooling at the same time as the heat dissipation, thereby preventing the SoCBox from becoming hot and enabling advanced calculations in the vehicle. It is expected that it can be used to cool not only the SoCBox but also the battery, which may become a solution to the high temperature caused by rapid charging.

[0361] Fig.25 An example of the system 10 is schematically shown. The system 10 includes a management server 100 . The system 10 includes a SoCBox 400 . The system 10 includes a cooling execution device 500 . The system 10 includes a cooling unit 600 .

[0362] SoCBox400, cooling execution device 500 and cooling unit 600 are mounted on the vehicle. SoCBox400 uses the sensor values ​​of multiple sensors mounted on the vehicle to control the automatic driving of the vehicle. The automatic driving control of the vehicle consumes a very high processing load, so there is a situation where SoCBox400 becomes very hot. If the SoCBox400 is too hot, it is possible that the operation of SoCBox400 cannot be performed normally, or it may have a bad effect on the vehicle.

[0363] The cooling execution device 500 involved in this embodiment predicts the temperature change of SoCBox400, and starts cooling SoCBox400 based on the temperature change. For example, the cooling execution device 500 responds to the prediction that SoCBox400 starts to heat up and immediately starts cooling SoCBox400. By starting cooling earlier than the start of heating or at the same time as the start of heating, it is possible to reliably prevent SoCBox400 from becoming high temperature. In addition, compared with the case where SoCBox400 is always cooled, the energy required for cooling can be reduced.

[0364] The cooling execution device 500 can predict the temperature change of the SoCBox 400 through AI. The learning of the temperature change of the SoCBox 400 can be performed using the data collected by the vehicle 200. For example, the management server 100 collects data from the vehicle 200 to perform learning. The subject that performs learning is not limited to the management server 100, and it can also be other devices.

[0365] The vehicle 200 is equipped with a SoCBox 400 and a temperature sensor 40 for measuring the temperature of the SoCBox 400. The SoCBox 400 controls the automatic driving of the vehicle 200 using the sensor values ​​of the multiple sensors mounted on the vehicle 200 and the external information received from the various servers 30. The server 30 may be an example of an external device. As examples of the various servers 30, a server that provides traffic information, a server that provides weather information, etc. may be cited. The SoCBox 400 sends the sensor values, external information, etc. used in the control of the automatic driving, and the temperature change of the SoCBox 400 during the control to the management server 100.

[0366] The management server 100 performs learning using information received from one or more SoCBox400. The management server 100 generates a learning model that uses the information obtained by SoCBox400 as input and the temperature change of SoCBox400 as output by performing machine learning using information such as sensor values ​​obtained by SoCBox400, external information, and the temperature change of SoCBox400 when SoCBox400 obtains such information as learning data.

[0367] The vehicle 300 is a vehicle having a cooling function according to the present embodiment. The vehicle 300 is equipped with a SoCBox 400, a cooling execution device 500, and a cooling unit 600. The cooling execution device 500 can receive a learning model generated by the management server 100 from the management server 100 and store the learning model.

[0368] The cooling execution device 500 obtains the sensor values ​​of multiple sensors mounted on the vehicle 300 obtained by the SoCBox400, the external information received from multiple servers 30, or obtains it from the SoCBox400, and inputs the obtained information into the learning model to predict the temperature change of the SoCBox400.

[0369] When it is predicted that the SoC Box 400 starts to generate heat and the temperature of the SoC Box 400 will become higher than a preset threshold, the cooling execution device 500 starts cooling the SoC Box 400 using the cooling unit 600 .

[0370] SoCBox400, cooling execution device 500, management server 100, and server 30 can communicate via network 20. Network 20 can include a vehicle network. Network 20 can include the Internet. Network 20 can include a LAN (Local Area Network). Network 20 can include a mobile communication network. The mobile communication network can comply with any of the communication methods after 5G (5th Generation) communication method, LTE (Long Term Evolution) communication method, 3G (3rd Generation) communication method, and 6G (6th Generation) communication method.

[0371] Fig.26 2 is an explanatory diagram for explaining the learning stage in the system 10. Here, as sensors 210 mounted on the vehicle 200, a camera 211, a LiDAR (Light Detection And Ranging) 212, a millimeter wave sensor 213, an ultrasonic sensor 214, an IMU sensor 215, and a GNSS (Global Navigation Satellite System) sensor 216 are exemplified. The vehicle 200 does not necessarily have to be equipped with all of these sensors, and may not have some of them, or may have sensors other than these sensors.

[0372] The SoCBox 400 acquires sensor information from each sensor included in the sensor 210. In addition, the SoCBox 400 can perform communication via the network 20, and the SoCBox 400 receives external information from a plurality of servers 30 respectively via the network 20. Then, the SoCBox 400 performs automatic driving control of the vehicle 200 using the acquired information.

[0373] The temperature sensor 40 measures the temperature change of the SoCBox 400. The SoCBox 400 transmits the sensor information received from the sensor 210, the external information received from the server 30, and the temperature change measured by the temperature sensor 40 when the automatic driving control is performed by acquiring these information to the management server 100.

[0374] The management server 100 includes an information acquisition unit 102 , a model generation unit 104 , and a model provision unit 106 . The information acquisition unit 102 acquires various information. The management server 100 can receive information transmitted from the SoCBox 400 .

[0375] The model generation unit 104 generates a learning model by performing machine learning using the information acquired by the information acquisition unit 102. The model generation unit 104 generates a learning model that uses the information acquired by SoCBox400 as input and the temperature change of SoCBox400 as output by performing machine learning using the information acquired by SoCBox400 and the temperature change of SoCBox400 when SoCBox400 acquires the information as learning data.

[0376] The model providing unit 106 provides the learning model generated by the model generating unit 104. The model providing unit 106 may transmit the learning model to the cooling execution device 500 mounted on the vehicle 300.

[0377] The system 10 may also be configured to predict the temperature change of each of the multiple parts of the SoCBox400. In this case, the vehicle 200 may be provided with multiple temperature sensors 40 that measure the temperature change of each of the multiple parts of the SoCBox400. The SoCBox400 may send the sensor information received from the sensor 210, the external information received from the server 30, and the temperature change measured by the multiple temperature sensors 40 when the automatic driving control is performed after obtaining the information to the management server 100. The model generation unit 104 performs machine learning by using the information obtained by the SoCBox400 and the temperature change of each of the multiple parts of the SoCBox400 when the SoCBox400 obtains the information as learning data, thereby generating a learning model that uses the information obtained by the SoCBox400 as input and the temperature change of each of the multiple parts of the SoCBox400 as output.

[0378] Fig. 27 This is an explanatory diagram for explaining the cooling execution stage in the system 10. Here, as sensors 310 mounted on the vehicle 300, a camera 311, a LiDAR 312, a millimeter wave sensor 313, an ultrasonic sensor 314, an IMU sensor 315, and a GNSS sensor 316 are exemplified. The vehicle 300 does not necessarily have to be equipped with all of these sensors, and may not have some of them, or may have sensors other than these sensors.

[0379] The cooling execution device 500 includes a detection unit 501, a model storage unit 502, an information acquisition unit 504, a prediction unit 506, and a cooling execution unit 508. The model storage unit 502 stores the learning model received from the management server 100. The information acquisition unit 504 acquires the information acquired by the SoCBox 400.

[0380] The detection unit 501 detects the temperature of a control device (SoCBox400) that controls the automatic driving of the vehicle and is mounted on the vehicle. In addition, the detection unit 501 can detect the temperature of each of multiple parts of the SoCBox400. For example, the detection unit 501 can detect the temperature of one or more SoCBox400, or detect the occurrence of a temperature rise or fall, or detect the value of the temperature change.

[0381] The information acquisition unit 504 acquires the sensor information acquired by the SoCBox400 from the sensor 310 from the sensor 310 or the SoCBox400. For example, the information acquisition unit 504 may receive the sensor information acquired by the SoCBox400 from the sensor 310 from the SoCBox400. The information acquisition unit 504 may also receive the same sensor information from the sensor 310 as the sensor information acquired by the SoCBox400 from the sensor 310. In this case, each of the sensors 310 may send the sensor information to the SoCBox400 and each of the cooling execution devices 500.

[0382] The information acquisition unit 504 acquires the external information acquired by the SoCBox400 from the server 30 from the server 30 or the SoCBox400. The information acquisition unit 504 may receive the external information received by the SoCBox400 from the server 30 from the SoCBox400. The information acquisition unit 504 may also receive the same external information from the server 30 as the external information received by the SoCBox400 from the server 30. In this case, the server 30 may send the external information to each of the SoCBox400 and the cooling execution device 500.

[0383] The prediction unit 506 predicts the temperature change of the control device (SoCBox400), which controls the automatic driving of the vehicle and is installed in the vehicle. The prediction unit 506 can predict the temperature change of SoCBox400 through AI. For example, the prediction unit 506 predicts the temperature change of SoCBox400 by inputting the information acquired by the information acquisition unit 504 into the learning model stored in the model storage unit 502. Furthermore, the prediction unit 506 can predict the temperature change of each of the multiple parts of the control device (SoCBox400). For example, the prediction unit 506 can predict the occurrence of a temperature increase or decrease for one or more SoCBox400, or predict the numerical value of the changed temperature.

[0384] The cooling execution unit 508 starts cooling the SoCBox 400 based on the temperature change of the SoCBox 400 predicted by the prediction unit 506. For example, the cooling execution unit 508 starts cooling the SoCBox 400 in response to the prediction unit 506 predicting that the SoCBox 400 starts to heat up. For example, the cooling execution unit 508 starts cooling the SoCBox 400 in response to the prediction unit 506 predicting that the temperature of the SoCBox 400 will become higher than a preset threshold.

[0385] The cooling execution unit 508 uses a given cooling unit to execute cooling of the control device (SoCBox400) based on the time that the temperature detected by the detection unit 501 continues to be above a given temperature. Specifically, when the time that the temperature continues to be above a given temperature exceeds a given threshold, the cooling execution unit 508 can use cooling units including one or more air cooling units, one or more water cooling units, and multiple liquid nitrogen cooling units to perform cooling. For example, when the temperature detected by the detection unit 501 continues to be above a given temperature for a time that exceeds a given threshold (for example, the time above the given temperature continues for more than 10 minutes, etc.), the cooling execution unit 508 can use a given cooling unit (for example, an air cooling unit, a water cooling unit, a liquid nitrogen cooling unit, etc.) to execute cooling of the control device. In addition, the aforementioned given threshold (time) related to the time that the temperature continues to be above a given temperature is a condition that can be set arbitrarily and is not limited.

[0386] In addition, the cooling execution unit 508 may perform rapid cooling when the temperature exceeds a given threshold. For example, the cooling execution unit 508 may perform rapid cooling when the temperature at which overheating occurs exceeds a given threshold. In addition, here, any threshold condition may be set for the so-called given threshold.

[0387] The cooling execution unit 508 may use the cooling unit 600 to execute cooling of the SoCBox 400. The cooling unit 600 may use an air cooling unit to cool the SoCBox 400. The cooling unit 600 may use a water cooling unit to cool the SoCBox 400. The cooling unit 600 may use a liquid nitrogen cooling unit to cool the SoCBox 400.

[0388] The cooling unit 600 may also include multiple cooling units. For example, the cooling unit 600 includes multiple air cooling units. For example, the cooling unit 600 includes multiple water cooling units. For example, the cooling unit 600 includes multiple liquid nitrogen cooling units. The cooling unit 600 may also include multiple of one or more air cooling units, one or more water cooling units, and one or more liquid nitrogen cooling units.

[0389] The plurality of cooling units may be configured to cool different parts of the SoCBox 400. The prediction unit 506 may use the information acquired by the information acquisition unit 504 to predict the temperature change of each of the plurality of parts of the SoCBox 400. The cooling execution unit 508 may start cooling the SoCBox 400 using a cooling unit selected from the plurality of cooling units that cool the plurality of parts of the SoCBox 400, based on the prediction result of the prediction unit 506.

[0390] The cooling execution unit 508 can use a cooling unit corresponding to the temperature of the SoCBox400 predicted by the prediction unit 506 to execute cooling of the SoCBox400. For example, the higher the temperature of the SoCBox400, the more cooling units the cooling execution unit 508 uses to execute cooling of the SoCBox400. As a specific example, when it is predicted that the temperature of the SoCBox400 will exceed a first threshold, the cooling execution unit 508 uses one of the plurality of cooling units to start cooling. Even so, when the temperature of the SoCBox400 still rises and is predicted to exceed a second threshold, the number of cooling units used is increased.

[0391] The cooling execution unit 508 can use a more powerful cooling unit to cool the SoCBox400 when the temperature of the SoCBox400 is higher. For example, when it is predicted that the temperature of the SoCBox400 will exceed the first threshold, the cooling execution unit 508 starts cooling using the air cooling unit. When the temperature of the SoCBox400 still rises despite this and is predicted to exceed the second threshold, the cooling execution unit 508 starts cooling using the water cooling unit. When the temperature of the SoCBox400 still rises despite this and is predicted to exceed the third threshold, the cooling execution unit 508 starts cooling using the liquid nitrogen cooling unit.

[0392] The cooling execution unit 508 executes cooling using the aforementioned cooling unit based on the time that the temperature remains above a given temperature as the temperature change predicted by the prediction unit 506. For example, when the vehicle is about to travel at a driving location where the temperature is predicted to remain above a given temperature for more than a given time as the temperature change, the cooling execution unit 508 can execute cooling using the aforementioned cooling unit.

[0393] The SoCBox 400 may have a plurality of processing chips, and the plurality of processing chips may be respectively arranged at different positions of the SoCBox 400. Each of the plurality of cooling units may be arranged at a position corresponding to each of the plurality of processing chips.

[0394] For example, when the number of processing chips used varies according to the control conditions of autonomous driving, cooling is performed using a cooling unit corresponding to the processing chips used, thereby enabling efficient cooling.

[0395] Fig.28 An example of the SoCBox 400 and the cooling unit 600 is schematically shown. Fig.28 The cooling unit 600 is illustrated as being composed of one cooling unit. When the cooling execution device 500 predicts that the SoCBox 400 starts to generate heat or that the temperature of the SoCBox 400 exceeds a preset threshold, the cooling unit 600 starts cooling, thereby cooling the entire SoCBox 400.

[0396] Fig.29 An example of the SoCBox 400 and the cooling unit 600 is schematically shown. Fig.29 The example shows a case where the cooling unit 600 is composed of a plurality of cooling units for cooling a plurality of parts of the SoCBox 400. The cooling execution device 500 predicts the temperature change of each of the plurality of parts of the SoCBox 400, and in response to the prediction that any part starts to generate heat or the temperature of any part will exceed a preset threshold, performs cooling using only the cooling unit corresponding to the part, thereby achieving efficient cooling.

[0397] Fig.30 An example of the SoCBox 400 and the cooling unit 600 is schematically shown. Fig.30The case where the cooling unit 600 is composed of two types of cooling units is illustrated. The cooling execution device 500 predicts the temperature change of each of the multiple parts of the SoCBox400, and in response to the prediction that any part begins to heat up or the temperature of any part will exceed a preset threshold, performs cooling using only the cooling unit corresponding to the part, thereby achieving efficient cooling. In addition, the cooling execution device 500 increases the cooling units used as the temperature of the SoCBox400 increases, that is, in this example, first, one of the two cooling units is used for cooling, and when the temperature of the SoCBox400 further increases, the other cooling unit is also used for cooling, thereby making the energy used for cooling more efficient.

[0398] <Sixth embodiment>

[0399] When the SoC (System on Chip) for autonomous driving performs advanced computing, heat generation becomes a technical problem. Therefore, in this embodiment, a technology called Synchronized Burst Chilling is provided that flexibly applies AI (Artificial Intelligence) to optimize the rapid cooling of the SoCBox.

[0400] The SoCBox will immediately become hot, making it difficult to perform advanced calculations in the vehicle (a technical issue for fully autonomous driving). AI predicts the heat dissipation of the SoCBox and performs cooling at the same time as the heat dissipation, thereby preventing the SoCBox from becoming hot and enabling advanced calculations in the vehicle. It is expected that it can be used to cool not only the SoCBox but also the battery, which may become a solution to the high temperature caused by rapid charging.

[0401] Fig.31 An example of the system 10 is schematically shown. The system 10 includes a management server 100 . The system 10 includes a SoCBox 400 . The system 10 includes a cooling execution device 500 . The system 10 includes a cooling unit 600 .

[0402] SoCBox400, cooling execution device 500 and cooling unit 600 are mounted on the vehicle. SoCBox400 uses the sensor values ​​of multiple sensors mounted on the vehicle to control the automatic driving of the vehicle. The automatic driving control of the vehicle consumes a very high processing load, so there is a situation where SoCBox400 becomes very hot. If the SoCBox400 is too hot, it is possible that the operation of SoCBox400 cannot be performed normally, or it may have a bad effect on the vehicle.

[0403] The cooling execution device 500 involved in this embodiment predicts the temperature change of SoCBox400, and starts cooling SoCBox400 based on the temperature change. For example, the cooling execution device 500 responds to the prediction that SoCBox400 starts to heat up and immediately starts cooling SoCBox400. By starting cooling earlier than the start of heating or at the same time as the start of heating, it is possible to reliably prevent SoCBox400 from becoming high temperature. In addition, compared with the case where SoCBox400 is always cooled, the energy required for cooling can be reduced.

[0404] The cooling execution device 500 can predict the temperature change of the SoCBox 400 through AI. The learning of the temperature change of the SoCBox 400 can be performed using the data collected by the vehicle 200. For example, the management server 100 collects data from the vehicle 200 to perform learning. The subject that performs learning is not limited to the management server 100, and it can also be other devices.

[0405] The vehicle 200 is equipped with a SoCBox 400 and a temperature sensor 40 for measuring the temperature of the SoCBox 400. The SoCBox 400 controls the automatic driving of the vehicle 200 using the sensor values ​​of the multiple sensors mounted on the vehicle 200 and the external information received from the various servers 30. The server 30 may be an example of an external device. As examples of the various servers 30, a server that provides traffic information, a server that provides weather information, etc. may be cited. The SoCBox 400 sends the sensor values, external information, etc. used in the control of the automatic driving, and the temperature change of the SoCBox 400 during the control to the management server 100.

[0406] The management server 100 performs learning using information received from one or more SoCBox400. The management server 100 generates a learning model that uses the information obtained by SoCBox400 as input and the temperature change of SoCBox400 as output by performing machine learning using information such as sensor values ​​obtained by SoCBox400, external information, and the temperature change of SoCBox400 when SoCBox400 obtains such information as learning data.

[0407] The vehicle 300 is a vehicle having a cooling function according to the present embodiment. The vehicle 300 is equipped with a SoCBox 400, a cooling execution device 500, and a cooling unit 600. The cooling execution device 500 can receive a learning model generated by the management server 100 from the management server 100 and store the learning model.

[0408] The cooling execution device 500 obtains the sensor values ​​of multiple sensors mounted on the vehicle 300 obtained by the SoCBox400, the external information received from multiple servers 30, or obtains it from the SoCBox400, and inputs the obtained information into the learning model to predict the temperature change of the SoCBox400.

[0409] When it is predicted that the SoC Box 400 starts to generate heat and the temperature of the SoC Box 400 will become higher than a preset threshold, the cooling execution device 500 starts cooling the SoC Box 400 using the cooling unit 600 .

[0410] SoCBox400, cooling execution device 500, management server 100, and server 30 can communicate via network 20. Network 20 can include a vehicle network. Network 20 can include the Internet. Network 20 can include a LAN (Local Area Network). Network 20 can include a mobile communication network. The mobile communication network can comply with any of the communication methods after 5G (5th Generation) communication method, LTE (Long Term Evolution) communication method, 3G (3rd Generation) communication method, and 6G (6th Generation) communication method.

[0411] Fig.32 2 is an explanatory diagram for explaining the learning stage in the system 10. Here, as sensors 210 mounted on the vehicle 200, a camera 211, a LiDAR (Light Detection And Ranging) 212, a millimeter wave sensor 213, an ultrasonic sensor 214, an IMU sensor 215, and a GNSS (Global Navigation Satellite System) sensor 216 are exemplified. The vehicle 200 does not necessarily have to be equipped with all of these sensors, and may not have some of them, or may have sensors other than these sensors.

[0412] The SoCBox 400 acquires sensor information from each sensor included in the sensor 210. In addition, the SoCBox 400 can perform communication via the network 20, and the SoCBox 400 receives external information from a plurality of servers 30 respectively via the network 20. Then, the SoCBox 400 performs automatic driving control of the vehicle 200 using the acquired information.

[0413] The temperature sensor 40 measures the temperature change of the SoCBox 400. The SoCBox 400 transmits the sensor information received from the sensor 210, the external information received from the server 30, and the temperature change measured by the temperature sensor 40 when the automatic driving control is performed by acquiring these information to the management server 100.

[0414] The management server 100 includes an information acquisition unit 102 , a model generation unit 104 , and a model provision unit 106 . The information acquisition unit 102 acquires various information. The management server 100 can receive information transmitted from the SoCBox 400 .

[0415] The model generation unit 104 generates a learning model by performing machine learning using the information acquired by the information acquisition unit 102. The model generation unit 104 generates a learning model that uses the information acquired by SoCBox400 as input and the temperature change of SoCBox400 as output by performing machine learning using the information acquired by SoCBox400 and the temperature change of SoCBox400 when SoCBox400 acquires the information as learning data.

[0416] The model providing unit 106 provides the learning model generated by the model generating unit 104. The model providing unit 106 may transmit the learning model to the cooling execution device 500 mounted on the vehicle 300.

[0417] The system 10 may also be configured to predict the temperature change of each of the multiple parts of the SoCBox400. In this case, the vehicle 200 may be provided with multiple temperature sensors 40 that measure the temperature change of each of the multiple parts of the SoCBox400. The SoCBox400 may send the sensor information received from the sensor 210, the external information received from the server 30, and the temperature change measured by the multiple temperature sensors 40 when the automatic driving control is performed after obtaining the information to the management server 100. The model generation unit 104 performs machine learning by using the information obtained by the SoCBox400 and the temperature change of each of the multiple parts of the SoCBox400 when the SoCBox400 obtains the information as learning data, thereby generating a learning model that uses the information obtained by the SoCBox400 as input and the temperature change of each of the multiple parts of the SoCBox400 as output.

[0418] Fig.33This is an explanatory diagram for explaining the cooling execution stage in the system 10. Here, as sensors 310 mounted on the vehicle 300, a camera 311, a LiDAR 312, a millimeter wave sensor 313, an ultrasonic sensor 314, an IMU sensor 315, and a GNSS sensor 316 are exemplified. The vehicle 300 does not necessarily have to be equipped with all of these sensors, and may not have some of them, or may have sensors other than these sensors.

[0419] The cooling execution device 500 includes a detection unit 501, a model storage unit 502, an information acquisition unit 504, an estimation unit 505, a prediction unit 506, a creation unit 507, and a cooling execution unit 508. The model storage unit 502 stores the learning model received from the management server 100. The information acquisition unit 504 acquires the information acquired by the SoCBox 400.

[0420] The detection unit 501 detects the temperature change of the control device (SoCBox400), which controls the automatic driving of the vehicle and is installed in the vehicle. In addition, the detection unit 501 can detect the temperature change of each of the multiple parts of the SoCBox400. For example, the detection unit 501 can detect the occurrence of a temperature rise or fall for one or more SoCBox400, or detect the value of the temperature change.

[0421] The estimation unit 505 estimates the switching between the automatic driving and the manual driving based on the preset driving route. Furthermore, the estimation unit 505 may estimate whether the switching from the automatic driving to the manual driving is to be generated based on the road conditions of the driving route. For example, the estimation unit 505 may estimate whether the switching from the automatic driving to the manual driving is to be generated based on the road conditions such as the large number of pedestrians, the occurrence of traffic jams, the occurrence of accidents, the occurrence of construction, the occurrence of prohibited passage, etc. for the preset driving route.

[0422] The information acquisition unit 504 acquires the sensor information acquired by the SoCBox400 from the sensor 310 from the sensor 310 or the SoCBox400. For example, the information acquisition unit 504 may receive the sensor information acquired by the SoCBox400 from the sensor 310 from the SoCBox400. The information acquisition unit 504 may also receive the same sensor information from the sensor 310 as the sensor information acquired by the SoCBox400 from the sensor 310. In this case, each of the sensors 310 may send the sensor information to the SoCBox400 and each of the cooling execution devices 500.

[0423] The information acquisition unit 504 acquires the external information acquired by the SoCBox400 from the server 30 from the server 30 or the SoCBox400. The information acquisition unit 504 may receive the external information received by the SoCBox400 from the server 30 from the SoCBox400. The information acquisition unit 504 may also receive the same external information from the server 30 as the external information received by the SoCBox400 from the server 30. In this case, the server 30 may send the external information to each of the SoCBox400 and the cooling execution device 500.

[0424] The prediction unit 506 predicts the temperature change of the control device (SoCBox400), which controls the automatic driving of the vehicle and is installed in the vehicle. The prediction unit 506 can predict the temperature change of the SoCBox400 through AI. For example, the prediction unit 506 predicts the temperature change of the SoCBox400 by inputting the information acquired by the information acquisition unit 504 into the learning model stored in the model storage unit 502.

[0425] The creation unit 507 creates a cooling plan with a given cooling unit based on the estimation result of the estimation unit 505. Specifically, when the estimation unit 505 estimates that the automatic driving is to be switched to the manual driving, the creation unit 507 may create a cooling plan with a cooling unit for a driving section of the manual driving that is assumed to generate more heat than the manual driving. On the other hand, when the estimation unit 505 estimates that the manual driving is to be switched to the automatic driving, the creation unit 507 may create a cooling plan with a cooling unit for a driving section of the automatic driving that is assumed to generate less heat than the automatic driving.

[0426] For example, when the estimation unit 505 estimates that the switch from automatic driving to manual driving will occur due to a change in road conditions on a preset driving route (a situation where automatic driving is difficult, etc.), the creation unit 507 can create a cooling plan with a cooling unit corresponding to manual driving. On the other hand, when the estimation unit 505 estimates that the switch from manual driving to automatic driving will occur due to a change in road conditions on a preset driving route (a situation where automatic driving is possible, etc.), the creation unit 507 can create a cooling plan with a cooling unit corresponding to automatic driving.

[0427] The creation unit 507 can compare the actual temperature change of the control device (SoCBox400) detected by the detection unit 501 with the predicted temperature change predicted by the prediction unit 506, and when the predicted temperature change is greater than the actual temperature change, create a cooling plan set with a given cooling unit. For example, the creation unit 507 can implement cooling using a given cooling unit based on a prediction result that the temperature change of the control device (SoCBox400) predicted by the prediction unit 506 is a temperature rise of 5°C and a detection result that the actual temperature change detected by the detection unit 501 is a temperature rise of 10°C.

[0428] The cooling execution unit 508 starts cooling the SoCBox 400 based on the temperature change of the SoCBox 400 predicted by the prediction unit 506. For example, the cooling execution unit 508 starts cooling the SoCBox 400 in response to the prediction unit 506 predicting that the SoCBox 400 starts to heat up. For example, the cooling execution unit 508 starts cooling the SoCBox 400 in response to the prediction unit 506 predicting that the temperature of the SoCBox 400 will become higher than a preset threshold.

[0429] The cooling execution unit 508 executes cooling of the control device (SoCBox400) which controls the automatic driving of the vehicle and is mounted on the vehicle based on the cooling plan created by the creation unit 507. For example, when the creation unit 507 creates a cooling plan for a driving interval of manual driving or a cooling plan for a driving interval of automatic driving, the cooling execution unit 508 can execute cooling of the control device (SoCBox400) with the cooling unit, cooling time, cooling timing, etc. set in the cooling plan.

[0430] In addition, when the creation unit 507 creates a cooling plan based on the comparison between the predicted temperature change and the actual temperature change, the cooling execution unit 508 can execute cooling of the control device (SoCBox400) with the cooling unit, cooling time, cooling timing, etc. set in the cooling plan.

[0431] The cooling execution unit 508 can execute cooling of the control device (SoCBox400) using a cooling unit selected from a plurality of cooling units for cooling each of the plurality of parts of the control device (SoCBox400) based on the cooling plan created by the creation unit 507. For example, in the case where an air cooling unit, a water cooling unit, and a liquid nitrogen cooling unit are set in the cooling plan created by the creation unit 507, and further the conditions for executing cooling of the plurality of parts of the control device are set, the cooling execution unit 508 can select one or more of the aforementioned plurality of cooling units to execute cooling.

[0432] The cooling execution unit 508 may use the cooling unit 600 to execute cooling of the SoCBox 400. The cooling unit 600 may use an air cooling unit to cool the SoCBox 400. The cooling unit 600 may use a water cooling unit to cool the SoCBox 400. The cooling unit 600 may use a liquid nitrogen cooling unit to cool the SoCBox 400.

[0433] The cooling unit 600 may also include multiple cooling units. For example, the cooling unit 600 includes multiple air cooling units. For example, the cooling unit 600 includes multiple water cooling units. For example, the cooling unit 600 includes multiple liquid nitrogen cooling units. The cooling unit 600 may also include multiple of one or more air cooling units, one or more water cooling units, and one or more liquid nitrogen cooling units.

[0434] The plurality of cooling units may be configured to cool different parts of the SoCBox 400. The prediction unit 506 may use the information acquired by the information acquisition unit 504 to predict the temperature change of each of the plurality of parts of the SoCBox 400. The cooling execution unit 508 may start cooling the SoCBox 400 using a cooling unit selected from the plurality of cooling units that cool the plurality of parts of the SoCBox 400, based on the prediction result of the prediction unit 506.

[0435] The cooling execution unit 508 can use a cooling unit corresponding to the temperature of the SoCBox400 predicted by the prediction unit 506 to execute cooling of the SoCBox400. For example, the higher the temperature of the SoCBox400, the more cooling units the cooling execution unit 508 uses to execute cooling of the SoCBox400. As a specific example, when it is predicted that the temperature of the SoCBox400 will exceed a first threshold, the cooling execution unit 508 uses one of the plurality of cooling units to start cooling. Even so, when the temperature of the SoCBox400 still rises and is predicted to exceed a second threshold, the number of cooling units used is increased.

[0436] The cooling execution unit 508 can use a more powerful cooling unit to cool the SoCBox400 when the temperature of the SoCBox400 is higher. For example, when it is predicted that the temperature of the SoCBox400 will exceed the first threshold, the cooling execution unit 508 starts cooling using the air cooling unit. When the temperature of the SoCBox400 still rises despite this and is predicted to exceed the second threshold, the cooling execution unit 508 starts cooling using the water cooling unit. When the temperature of the SoCBox400 still rises despite this and is predicted to exceed the third threshold, the cooling execution unit 508 starts cooling using the liquid nitrogen cooling unit.

[0437] The SoCBox 400 may have a plurality of processing chips, and the plurality of processing chips may be respectively arranged at different positions of the SoCBox 400. Each of the plurality of cooling units may be arranged at a position corresponding to each of the plurality of processing chips.

[0438] For example, when the number of processing chips used varies according to the control conditions of autonomous driving, cooling is performed using a cooling unit corresponding to the processing chips used, thereby enabling efficient cooling.

[0439] Fig.34 An example of the SoCBox 400 and the cooling unit 600 is schematically shown. Fig.34 The cooling unit 600 is illustrated as being composed of one cooling unit. When the cooling execution device 500 predicts that the SoCBox 400 starts to generate heat or that the temperature of the SoCBox 400 exceeds a preset threshold, the cooling unit 600 starts cooling, thereby cooling the entire SoCBox 400.

[0440] Fig.35 An example of the SoCBox 400 and the cooling unit 600 is schematically shown. Fig.35 The example shows a case where the cooling unit 600 is composed of a plurality of cooling units for cooling a plurality of parts of the SoCBox 400. The cooling execution device 500 predicts the temperature change of each of the plurality of parts of the SoCBox 400, and in response to the prediction that any part starts to generate heat or the temperature of any part will exceed a preset threshold, performs cooling using only the cooling unit corresponding to the part, thereby achieving efficient cooling.

[0441] Fig.36 An example of the SoCBox 400 and the cooling unit 600 is schematically shown. Fig.36 The case where the cooling unit 600 is composed of two types of cooling units is illustrated. The cooling execution device 500 predicts the temperature change of each of the multiple parts of the SoCBox400, and in response to the prediction that any part begins to heat up or the temperature of any part will exceed a preset threshold, performs cooling using only the cooling unit corresponding to the part, thereby achieving efficient cooling. In addition, the cooling execution device 500 increases the cooling units used as the temperature of the SoCBox400 increases, that is, in this example, first, one of the two cooling units is used for cooling, and when the temperature of the SoCBox400 further increases, the other cooling unit is also used for cooling, thereby making the energy used for cooling more efficient.

[0442] <Seventh embodiment>

[0443] When the SoC (System on Chip) (processing chip) for autonomous driving performs advanced computing, heat generation becomes a technical problem. Therefore, in this embodiment, a technology called Synchronized Burst Chilling is provided that flexibly applies AI (Artificial Intelligence) to optimize the rapid cooling of the SoCBox.

[0444] The SoCBox will immediately reach a high temperature, making it difficult to perform advanced calculations in the vehicle (a technical issue for fully autonomous driving). For example, the SoCBox has multiple SoCs, and if the control of autonomous driving becomes complicated, the processing performed by the SoC will increase, so there is a situation where the SoCBox will reach a high temperature and overheat. Therefore, AI predicts the heat dissipation of the SoCBox and performs cooling to keep the temperature of the SoCBox within a given temperature range, thereby preventing the SoCBox from reaching a high temperature and making advanced calculations in the vehicle possible. It is expected that it can be used not only for cooling the SoCBox, but also for cooling the battery, which may become a solution to the high temperature caused by rapid charging.

[0445] Fig.37 An example of the system 10 is schematically shown. The system 10 includes a management server 100 . The system 10 includes a SoCBox 400 . The system 10 includes a cooling device 700 . The system 10 includes a cooling unit 600 .

[0446] SoCBox400, cooling device 700 and cooling unit 600 are mounted on the vehicle. SoCBox400 is a control device having multiple SoCs, and uses the sensor values ​​of multiple sensors mounted on the vehicle to control the automatic driving of the vehicle. The automatic driving control of the vehicle consumes a very high processing load, so there is a situation where SoCBox400 becomes very hot. If the SoCBox400 is too hot, it is possible that the operation of SoCBox400 cannot be performed normally, or it may have a negative impact on the vehicle.

[0447] The cooling device 700 involved in this embodiment predicts the temperature change of SoCBox400, and based on the temperature change, cools SoCBox400 to keep the temperature of SoCBox400 within a given temperature range. For example, the cooling device 700 predicts that SoCBox400 starts to heat up, and when the predicted temperature after heating exceeds the given temperature range, the SoCBox400 is cooled to keep it within the given temperature range.

[0448] Thus, the cooling device 700 can maintain the temperature of the SoCBox 400 within a given temperature range, so the temperature of the SoCBox 400 is rarely rapidly cooled, and as a result, the total energy consumed for cooling can be reduced.

[0449] The cooling device 700 may also determine a given temperature range in accordance with a threshold temperature pre-set for the SoCBox400. For example, the cooling device 700 determines a temperature range around an arbitrary temperature about 5°C lower than a threshold temperature pre-set in consideration of normal execution of various SoC processes in the SoCBox400 as a given temperature range.

[0450] Thus, the cooling device 700 always cools the SoCBox400 to keep it within an arbitrary temperature range of about 5°C lower than the threshold temperature, rather than performing the process of rapidly starting cooling several times after the predicted temperature after heat generation becomes around the threshold temperature, thereby reducing the total energy used for cooling.

[0451] Here, the given temperature range refers to a temperature range centered on an arbitrary temperature lower than a threshold temperature set as the upper limit temperature at which multiple SoCs possessed by the SoCBox can normally process, and there is no limitation on the temperature difference between the threshold temperature and the temperature at the center of the temperature range, or the width of the temperature range.

[0452] The cooling device 700 can predict the temperature change of the SoCBox 400 through AI. In addition, the learning of the temperature change of the SoCBox 400 can be performed using the data collected by the vehicle 200. For example, the management server 100 collects data from the vehicle 200 to perform learning. The subject that performs learning is not limited to the management server 100, and it can also be other devices.

[0453] The vehicle 200 is equipped with a SoCBox 400 and a temperature sensor 40 for measuring the temperature of the SoCBox 400. The SoCBox 400 controls the automatic driving of the vehicle 200 using sensor values ​​of a plurality of sensors mounted on the vehicle 200 and external information received from a variety of servers 30.

[0454] The server 30 may be an example of an external device. Examples of various servers 30 include a server that provides traffic information, a server that provides weather information, etc. The SoCBox 400 sends sensor values ​​used in the control of autonomous driving, external information, etc., and temperature changes of the SoCBox 400 during control to the management server 100.

[0455] The management server 100 performs learning using information received from one or more SoCBox400. The management server 100 generates a learning model that uses the information obtained by SoCBox400 as input and the temperature change of SoCBox400 as output by performing machine learning using information such as sensor values ​​obtained by SoCBox400, external information, and the temperature change of SoCBox400 when SoCBox400 obtains such information as learning data.

[0456] The vehicle 300 is a vehicle having a cooling function according to the present embodiment. The SoCBox 400, the cooling device 700, and the cooling unit 600 are mounted on the vehicle 300. The cooling device 700 can receive the learning model generated by the management server 100 from the management server 100 and store the learning model.

[0457] The cooling device 700 obtains sensor values ​​of multiple sensors mounted on the vehicle 300 obtained by the SoCBox400, external information received from multiple servers 30, or from the SoCBox400, and inputs the obtained information into the learning model to predict the temperature change of the SoCBox400.

[0458] The cooling device 700 predicts that the SoC Box 400 starts to generate heat, and when the predicted temperature after the heat generation exceeds a predetermined temperature range, the cooling unit 600 starts cooling the SoC Box 400 to maintain the temperature of the SoC Box 400 within the predetermined temperature range.

[0459] The SoC Box 400, the cooling device 700, the management server 100, and the server 30 can communicate via the network 20. The network 20 can include a vehicle network. The network 20 can include the Internet. The network 20 can include a LAN (Local Area Network).

[0460] The network 20 may include a mobile communication network. The mobile communication network may comply with any of the 5G (5th Generation) communication method, the LTE (Long Term Evolution) communication method, the 3G (3rd Generation) communication method, and the 6G (6th Generation) communication method or later communication methods.

[0461] Fig.382 is an explanatory diagram for explaining the learning stage in the system 10. Here, as sensors 210 mounted on the vehicle 200, a camera 211, a LiDAR (Light Detection And Ranging) 212, a millimeter wave sensor 213, an ultrasonic sensor 214, an IMU sensor 215, and a GNSS (Global Navigation Satellite System) sensor 216 are exemplified. The vehicle 200 does not necessarily have to be equipped with all of these sensors, and may not have some of them, or may have sensors other than these sensors.

[0462] The SoCBox 400 acquires sensor information from each sensor included in the sensor 210. In addition, the SoCBox 400 can perform communication via the network 20, and the SoCBox 400 receives external information from a plurality of servers 30 respectively via the network 20. Then, the SoCBox 400 performs automatic driving control of the vehicle 200 using the acquired information.

[0463] The temperature sensor 40 measures the temperature change of the SoCBox 400. The SoCBox 400 transmits the sensor information received from the sensor 210, the external information received from the server 30, and the temperature change measured by the temperature sensor 40 when the automatic driving control is performed by acquiring these information to the management server 100.

[0464] The management server 100 includes an information acquisition unit 102 , a model generation unit 104 , and a model provision unit 106 . The information acquisition unit 102 acquires various information. The management server 100 can receive information transmitted from the SoCBox 400 .

[0465] The model generation unit 104 generates a learning model by performing machine learning using the information acquired by the information acquisition unit 102. The model generation unit 104 generates a learning model that uses the information acquired by SoCBox400 as input and the temperature change of SoCBox400 as output by performing machine learning using the information acquired by SoCBox400 and the temperature change of SoCBox400 when SoCBox400 acquires the information as learning data.

[0466] The model providing unit 106 provides the learning model generated by the model generating unit 104. The model providing unit 106 may transmit the learning model to the cooling device 700 mounted on the vehicle 300.

[0467] The system 10 may also be configured to predict the temperature change of each of the multiple parts of the SoCBox400. In this case, the vehicle 200 may be provided with multiple temperature sensors 40 for measuring the temperature change of each of the multiple parts of the SoCBox400. The SoCBox400 may send the sensor information received from the sensor 210, the external information received from the server 30, and the temperature change measured by the multiple temperature sensors 40 when the automatic driving control is performed by acquiring these information to the management server 100.

[0468] The model generation unit 104 performs machine learning by using the information acquired by SoCBox400 and the temperature changes of multiple parts of SoCBox400 when SoCBox400 acquires the information as learning data, thereby generating a learning model that uses the information acquired by SoCBox400 as input and the temperature changes of multiple parts of SoCBox400 as output.

[0469] Fig.39 This is an explanatory diagram for explaining the cooling execution stage in the system 10. Here, as sensors 310 mounted on the vehicle 300, a camera 311, a LiDAR 312, a millimeter wave sensor 313, an ultrasonic sensor 314, an IMU sensor 315, and a GNSS sensor 316 are exemplified. The vehicle 300 does not necessarily have to be equipped with all of these sensors, and may not have some of them, or may have sensors other than these sensors.

[0470] The cooling device 700 includes a model storage unit 502, an information acquisition unit 504, a prediction unit 506, and a cooling execution unit 508. The model storage unit 502 stores the learning model received from the management server 100. The information acquisition unit 504 acquires the information acquired by the SoCBox 400.

[0471] The information acquisition unit 504 acquires the sensor information acquired by the SoCBox 400 from the sensor 310 from the sensor 310 or the SoCBox 400. For example, the information acquisition unit 504 may receive the sensor information acquired by the SoCBox 400 from the sensor 310 from the SoCBox 400.

[0472] The information acquisition section 504 may also receive the same sensor information as the sensor information acquired by the SoCBox 400 from the sensor 310 from the sensor 310. In this case, each of the sensors 310 may transmit the sensor information to each of the SoCBox 400 and the cooling device 700.

[0473] The information acquisition unit 504 acquires the external information acquired by the SoCBox400 from the server 30 from the server 30 or the SoCBox400. The information acquisition unit 504 may receive the external information received by the SoCBox400 from the server 30 from the SoCBox400. The information acquisition unit 504 may also receive the same external information from the server 30 as the external information received by the SoCBox400 from the server 30. In this case, the server 30 may send the external information to each of the SoCBox400 and the cooling device 700.

[0474] The prediction unit 506 predicts the temperature change of the SoCBox 400. The prediction unit 506 may predict the temperature change of the SoCBox 400 through AI. For example, the prediction unit 506 predicts the temperature change of the SoCBox 400 by inputting the information acquired by the information acquisition unit 504 into the learning model stored in the model storage unit 502.

[0475] The cooling execution unit 508 performs cooling to keep the temperature of the SoCBox 400 within a given temperature range based on the temperature change of the SoCBox 400 predicted by the prediction unit 506. For example, in response to the situation that the temperature of the SoCBox 400 predicted by the prediction unit 506 will become higher than the upper limit temperature of the given temperature range, the cooling execution unit 508 starts cooling the SoCBox 400 to keep the temperature of the SoCBox 400 within the given temperature range.

[0476] The cooling execution unit 508 may use the cooling unit 600 to execute cooling of the SoCBox 400. The cooling unit 600 may use an air cooling unit to cool the SoCBox 400. The cooling unit 600 may use a water cooling unit to cool the SoCBox 400. The cooling unit 600 may use a liquid nitrogen cooling unit to cool the SoCBox 400.

[0477] The cooling unit 600 may also include multiple cooling units. For example, the cooling unit 600 includes multiple air cooling units. For example, the cooling unit 600 includes multiple water cooling units. For example, the cooling unit 600 includes multiple liquid nitrogen cooling units. The cooling unit 600 may also include multiple of one or more air cooling units, one or more water cooling units, and one or more liquid nitrogen cooling units.

[0478] The plurality of cooling units may be configured to cool different parts of the SoCBox 400. The prediction unit 506 may use the information acquired by the information acquisition unit 504 to predict the temperature change of each of the plurality of parts of the SoCBox 400. The cooling execution unit 508 may start cooling the SoCBox 400 using a cooling unit selected from the plurality of cooling units that cool the plurality of parts of the SoCBox 400, based on the prediction result of the prediction unit 506.

[0479] The cooling execution unit 508 may cool the SoC Box 400 using cooling units corresponding to the temperature of the SoC Box 400 predicted by the prediction unit 506. For example, the higher the temperature of the SoC Box 400, the more cooling units the cooling execution unit 508 uses to cool the SoC Box 400.

[0480] As a specific example, when it is predicted that the temperature of SoCBox400 will exceed the upper limit temperature within a given temperature range, the cooling execution unit 508 starts cooling using one of the multiple cooling units. Even so, if the temperature of SoCBox400 still rises and is predicted to exceed the aforementioned upper limit temperature, the number of cooling units used is increased.

[0481] The cooling execution unit 508 can use a more powerful cooling unit to cool the SoCBox400 when the temperature of the SoCBox400 is higher. For example, when it is predicted that the temperature of the SoCBox400 will exceed the aforementioned upper limit temperature, the cooling execution unit 508 starts cooling using the air cooling unit. Even so, if the temperature of the SoCBox400 still rises and is predicted to exceed the aforementioned upper limit temperature, the cooling execution unit 508 starts cooling using the water cooling unit. Even so, if the temperature of the SoCBox400 still rises and is predicted to exceed the aforementioned upper limit temperature, the cooling execution unit 508 starts cooling using the liquid nitrogen cooling unit.

[0482] The SoCBox 400 may have a plurality of processing chips, and the plurality of processing chips may be respectively arranged at different positions of the SoCBox 400. Each of the plurality of cooling units may be arranged at a position corresponding to each of the plurality of processing chips.

[0483] For example, when the number of processing chips used varies according to the control conditions of autonomous driving, cooling is performed using a cooling unit corresponding to the processing chips used, thereby enabling efficient cooling.

[0484] Fig.40 An example of the SoCBox 400 and the cooling unit 600 is schematically shown. Fig.40 The cooling unit 600 is shown as a single cooling unit. When the cooling device 700 predicts that the SoCBox 400 starts to generate heat or that the temperature of the SoCBox 400 will exceed a predetermined temperature range, the cooling unit 600 starts cooling, thereby cooling the entire SoCBox 400.

[0485] Fig.41An example of the SoCBox 400 and the cooling unit 600 is schematically shown. Fig.41 The case where the cooling unit 600 is composed of a plurality of cooling units that cool a plurality of locations of the SoCBox 400 , respectively, is illustrated.

[0486] The cooling device 700 predicts the temperature changes of each of the multiple parts of the SoCBox400, and in response to the prediction that any part starts to heat up or the temperature of any part will exceed a given temperature range, performs cooling using only the cooling unit corresponding to the part, thereby achieving efficient cooling.

[0487] Fig.42 An example of the SoCBox 400 and the cooling unit 600 is schematically shown. Fig.42 The case where the cooling unit 600 is composed of two types of cooling units is illustrated. The cooling device 700 predicts the temperature change of each of the multiple parts of the SoCBox 400, and in response to the prediction that any part starts to heat up or the temperature of any part will exceed a given temperature range, performs cooling using only the cooling unit corresponding to the part, thereby achieving efficient cooling.

[0488] In addition, the cooling device 700 increases the number of cooling units used as the temperature of the SoCBox 400 increases. That is, in this example, first, one of the two cooling units is used for cooling, and when the temperature of the SoCBox 400 further increases, the other cooling unit is also used for cooling, thereby making the energy used for cooling more efficient.

[0489] <Eighth Embodiment>

[0490] When the SoC (System on Chip) for autonomous driving performs advanced computing, heat generation becomes a technical problem. Therefore, in this embodiment, a technology called Synchronized Burst Chilling is provided that flexibly applies AI (Artificial Intelligence) to optimize the rapid cooling of the SoCBox.

[0491] The SoCBox will immediately become hot, making it difficult to perform advanced calculations in the vehicle (a technical issue for fully autonomous driving). AI predicts the heat dissipation of the SoCBox and performs cooling at the same time as the heat dissipation, thereby preventing the SoCBox from becoming hot and enabling advanced calculations in the vehicle. It is expected that it can be used to cool not only the SoCBox but also the battery, which may become a solution to the high temperature caused by rapid charging.

[0492] Fig.43An example of the system 10 is schematically shown. The system 10 includes a management server 100 . The system 10 includes a SoCBox 400 . The system 10 includes a cooling execution device 500 . The system 10 includes a cooling unit 600 .

[0493] SoCBox400, cooling execution device 500 and cooling unit 600 are mounted on the vehicle. SoCBox400 uses the sensor values ​​of multiple sensors mounted on the vehicle to control the automatic driving of the vehicle. The automatic driving control of the vehicle consumes a very high processing load, so there is a situation where SoCBox400 becomes very hot. If the SoCBox400 is too hot, it is possible that the operation of SoCBox400 cannot be performed normally, or it may have a bad effect on the vehicle.

[0494] The cooling execution device 500 involved in this embodiment predicts the temperature change of SoCBox400, and starts cooling SoCBox400 based on the temperature change. For example, the cooling execution device 500 responds to the prediction that SoCBox400 starts to heat up and immediately starts cooling SoCBox400. By starting cooling earlier than the start of heating or at the same time as the start of heating, it is possible to reliably prevent SoCBox400 from becoming high temperature. In addition, compared with the case where SoCBox400 is always cooled, the energy required for cooling can be reduced.

[0495] The cooling execution device 500 can predict the temperature change of the SoCBox 400 through AI. The learning of the temperature change of the SoCBox 400 can be performed using the data collected by the vehicle 200. For example, the management server 100 collects data from the vehicle 200 to perform learning. The subject that performs learning is not limited to the management server 100, and it can also be other devices.

[0496] The vehicle 200 is equipped with a SoCBox 400 and a temperature sensor 40 for measuring the temperature of the SoCBox 400. The SoCBox 400 controls the automatic driving of the vehicle 200 using the sensor values ​​of the multiple sensors mounted on the vehicle 200 and the external information received from the various servers 30. The server 30 may be an example of an external device. As examples of the various servers 30, a server that provides traffic information, a server that provides weather information, etc. may be cited. The SoCBox 400 sends the sensor values, external information, etc. used in the control of the automatic driving, and the temperature change of the SoCBox 400 during the control to the management server 100.

[0497] The management server 100 performs learning using information received from one or more SoCBox400. The management server 100 generates a learning model that uses the information obtained by SoCBox400 as input and the temperature change of SoCBox400 as output by performing machine learning using information such as sensor values ​​obtained by SoCBox400, external information, and the temperature change of SoCBox400 when SoCBox400 obtains such information as learning data.

[0498] The vehicle 300 is a vehicle having a cooling function according to the present embodiment. The vehicle 300 is equipped with a SoCBox 400, a cooling execution device 500, and a cooling unit 600. The cooling execution device 500 can receive a learning model generated by the management server 100 from the management server 100 and store the learning model.

[0499] The cooling execution device 500 obtains the sensor values ​​of multiple sensors mounted on the vehicle 300 obtained by the SoCBox400, the external information received from multiple servers 30, or obtains it from the SoCBox400, and inputs the obtained information into the learning model to predict the temperature change of the SoCBox400.

[0500] When it is predicted that the SoC Box 400 starts to generate heat and the temperature of the SoC Box 400 will become higher than a preset threshold, the cooling execution device 500 starts cooling the SoC Box 400 using the cooling unit 600 .

[0501] SoCBox400, cooling execution device 500, management server 100, and server 30 can communicate via network 20. Network 20 can include a vehicle network. Network 20 can include the Internet. Network 20 can include a LAN (Local Area Network). Network 20 can include a mobile communication network. The mobile communication network can comply with any of the communication methods after 5G (5th Generation) communication method, LTE (Long Term Evolution) communication method, 3G (3rd Generation) communication method, and 6G (6th Generation) communication method.

[0502] Fig.442 is an explanatory diagram for explaining the learning stage in the system 10. Here, as sensors 210 mounted on the vehicle 200, a camera 211, a LiDAR (Light Detection And Ranging) 212, a millimeter wave sensor 213, an ultrasonic sensor 214, an IMU sensor 215, and a GNSS (Global Navigation Satellite System) sensor 216 are exemplified. The vehicle 200 does not necessarily have to be equipped with all of these sensors, and may not have some of them, or may have sensors other than these sensors.

[0503] The SoCBox 400 acquires sensor information from each sensor included in the sensor 210. In addition, the SoCBox 400 can perform communication via the network 20, and the SoCBox 400 receives external information from a plurality of servers 30 respectively via the network 20. Then, the SoCBox 400 performs automatic driving control of the vehicle 200 using the acquired information.

[0504] The temperature sensor 40 measures the temperature change of the SoCBox 400. The SoCBox 400 transmits the sensor information received from the sensor 210, the external information received from the server 30, and the temperature change measured by the temperature sensor 40 when the automatic driving control is performed by acquiring these information to the management server 100.

[0505] The management server 100 includes an information acquisition unit 102 , a model generation unit 104 , and a model provision unit 106 . The information acquisition unit 102 acquires various information. The management server 100 can receive information transmitted from the SoCBox 400 .

[0506] The model generation unit 104 generates a learning model by performing machine learning using the information acquired by the information acquisition unit 102. The model generation unit 104 generates a learning model that uses the information acquired by SoCBox400 as input and the temperature change of SoCBox400 as output by performing machine learning using the information acquired by SoCBox400 and the temperature change of SoCBox400 when SoCBox400 acquires the information as learning data.

[0507] The model providing unit 106 provides the learning model generated by the model generating unit 104. The model providing unit 106 may transmit the learning model to the cooling execution device 500 mounted on the vehicle 300.

[0508] The system 10 may also be configured to predict the temperature change of each of the multiple parts of the SoCBox400. In this case, the vehicle 200 may be provided with multiple temperature sensors 40 that measure the temperature change of each of the multiple parts of the SoCBox400. The SoCBox400 may send the sensor information received from the sensor 210, the external information received from the server 30, and the temperature change measured by the multiple temperature sensors 40 when the automatic driving control is performed after obtaining the information to the management server 100. The model generation unit 104 performs machine learning by using the information obtained by the SoCBox400 and the temperature change of each of the multiple parts of the SoCBox400 when the SoCBox400 obtains the information as learning data, thereby generating a learning model that uses the information obtained by the SoCBox400 as input and the temperature change of each of the multiple parts of the SoCBox400 as output.

[0509] Fig.45 This is an explanatory diagram for explaining the cooling execution stage in the system 10. Here, as sensors 310 mounted on the vehicle 300, a camera 311, a LiDAR 312, a millimeter wave sensor 313, an ultrasonic sensor 314, an IMU sensor 315, and a GNSS sensor 316 are exemplified. The vehicle 300 does not necessarily have to be equipped with all of these sensors, and may not have some of them, or may have sensors other than these sensors.

[0510] The cooling execution device 500 includes a detection unit 501, a model storage unit 502, an information acquisition unit 504, a prediction unit 506, a selection unit 510, a cooling execution unit 508, and an output unit 509. The model storage unit 502 stores the learning model received from the management server 100. The information acquisition unit 504 acquires information acquired by the SoCBox 400.

[0511] The detection unit 501 detects a temperature change of a control device that controls the automatic driving of the vehicle and is mounted on the vehicle. In addition, the detection unit 501 can detect a temperature change of each of multiple parts of the SoCBox400. For example, the detection unit 501 can detect the occurrence of a temperature rise or fall for one or more SoCBox400, or detect the value of the temperature change.

[0512] The information acquisition unit 504 acquires the sensor information acquired by the SoCBox400 from the sensor 310 from the sensor 310 or the SoCBox400. For example, the information acquisition unit 504 may receive the sensor information acquired by the SoCBox400 from the sensor 310 from the SoCBox400. The information acquisition unit 504 may also receive the same sensor information from the sensor 310 as the sensor information acquired by the SoCBox400 from the sensor 310. In this case, each of the sensors 310 may send the sensor information to the SoCBox400 and each of the cooling execution devices 500.

[0513] The information acquisition unit 504 acquires the external information acquired by SoCBox400 from the server 30 from the server 30 or SoCBox400. The information acquisition unit 504 can receive the external information received by SoCBox400 from the server 30 from the SoCBox400. The information acquisition unit 504 can also receive the same external information as the external information received by SoCBox400 from the server 30 from the server 30. In this case, the server 30 can send the external information to each of the SoCBox400 and the cooling execution device 500. In addition, the information acquisition unit 504 can acquire the result of the information processing involved in the automatic driving from the external information processing device, which performs the information processing involved in the automatic driving as the given information processing of the control device (SoCBox400). For example, when the external information processing device implements the information processing involved in the control of the automatic driving that should have been implemented by the control device (SoCBox400) of the vehicle, the information acquisition unit 504 can acquire the result of the information processing involved in the automatic driving from the external information processing device.

[0514] The prediction unit 506 predicts the temperature change of the SoCBox 400. The prediction unit 506 may predict the temperature change of the SoCBox 400 through AI. For example, the prediction unit 506 predicts the temperature change of the SoCBox 400 by inputting the information acquired by the information acquisition unit 504 into the learning model stored in the model storage unit 502.

[0515] When the temperature change exceeds a given threshold value, the selection unit 510 selects a given operating condition for lowering the temperature of the control device. From now on, a specific example of the operating condition selected by the selection unit 510 will be described.

[0516] For example, when the temperature change exceeds a given threshold, the selection unit 510 can select an operating condition that suppresses the amount of calculation involved in a given information processing of the control device (SoCBox400) as a given operating condition. Thus, the cooling execution device suppresses the heat generated by the calculation involved in the information processing, and has a cooling effect on the control device.

[0517] When the temperature change exceeds a given threshold, the selection unit 510 may select an operating condition that suppresses the driving speed of the automatic driving to a given speed or less as the given operating condition. Thus, the cooling execution device suppresses the amount of calculation involved in the information processing of the automatic driving, thereby suppressing the heat generated by the calculation, and thus achieving a cooling effect on the control device.

[0518] In addition, when the temperature change exceeds a given threshold, the selection unit 510 may select an operating condition for changing the automatic driving to the manual driving as the given operating condition. Thus, the cooling execution device suppresses the heat generated by the calculation by switching to the manual driving which has less calculation amount than the automatic driving, thereby achieving a cooling effect on the control device.

[0519] When the temperature change exceeds a given threshold, the selection unit 510 may select an operation condition of stopping at a given position (e.g., a shoulder of a road) of a road included in the driving route of the automatic driving as the given operation condition. Thus, the cooling execution device suppresses the amount of calculation involved in the automatic driving by stopping the automatic driving, thereby suppressing the heat generated by the calculation, and achieving a cooling effect on the control device.

[0520] When the temperature change exceeds a given threshold, the selection unit 510 may select an operating condition that suppresses the acquisition of given information used in the information processing of the autonomous driving as the given operating condition. Thus, the cooling execution device suppresses the acquisition of information used in the information processing of the autonomous driving, suppresses the amount of calculation involved in the autonomous driving, thereby suppressing the heat generated by the calculation, and has a cooling effect on the control device.

[0521] The cooling execution unit 508 starts cooling the SoCBox 400 based on the temperature change of the SoCBox 400 predicted by the prediction unit 506. For example, the cooling execution unit 508 starts cooling the SoCBox 400 in response to the prediction unit 506 predicting that the SoCBox 400 starts to heat up. For example, the cooling execution unit 508 starts cooling the SoCBox 400 in response to the prediction unit 506 predicting that the temperature of the SoCBox 400 will become higher than a preset threshold.

[0522] The cooling execution unit 508 may use the cooling unit 600 to execute cooling of the SoCBox 400. The cooling unit 600 may use an air cooling unit to cool the SoCBox 400. The cooling unit 600 may use a water cooling unit to cool the SoCBox 400. The cooling unit 600 may use a liquid nitrogen cooling unit to cool the SoCBox 400.

[0523] The cooling unit 600 may also include multiple cooling units. For example, the cooling unit 600 includes multiple air cooling units. For example, the cooling unit 600 includes multiple water cooling units. For example, the cooling unit 600 includes multiple liquid nitrogen cooling units. The cooling unit 600 may also include multiple of one or more air cooling units, one or more water cooling units, and one or more liquid nitrogen cooling units.

[0524] The plurality of cooling units may be configured to cool different parts of the SoCBox 400. The prediction unit 506 may use the information acquired by the information acquisition unit 504 to predict the temperature change of each of the plurality of parts of the SoCBox 400. The cooling execution unit 508 may start cooling the SoCBox 400 using a cooling unit selected from the plurality of cooling units that cool the plurality of parts of the SoCBox 400, based on the prediction result of the prediction unit 506.

[0525] The cooling execution unit 508 can use a cooling unit corresponding to the temperature of the SoCBox400 predicted by the prediction unit 506 to execute cooling of the SoCBox400. For example, the higher the temperature of the SoCBox400, the more cooling units the cooling execution unit 508 uses to execute cooling of the SoCBox400. As a specific example, when it is predicted that the temperature of the SoCBox400 will exceed a first threshold, the cooling execution unit 508 uses one of the plurality of cooling units to start cooling. Even so, when the temperature of the SoCBox400 still rises and is predicted to exceed a second threshold, the number of cooling units used is increased.

[0526] The cooling execution unit 508 can use a more powerful cooling unit to cool the SoCBox400 when the temperature of the SoCBox400 is higher. For example, when it is predicted that the temperature of the SoCBox400 will exceed the first threshold, the cooling execution unit 508 starts cooling using the air cooling unit. When the temperature of the SoCBox400 still rises despite this and is predicted to exceed the second threshold, the cooling execution unit 508 starts cooling using the water cooling unit. When the temperature of the SoCBox400 still rises despite this and is predicted to exceed the third threshold, the cooling execution unit 508 starts cooling using the liquid nitrogen cooling unit.

[0527] The output unit 509 outputs a given operating condition based on the selection result of the selection unit 510. Specifically, the output unit 509 can output information related to the operating condition for the control device (SoCBox400) to control the automatic driving of the vehicle based on the operating condition selected by the selection unit 510. In addition, the information related to the operating condition output by the output unit 509 is not particularly limited as long as it is in a form that can be used by the vehicle control device.

[0528] The SoCBox 400 may have a plurality of processing chips, and the plurality of processing chips may be respectively arranged at different positions of the SoCBox 400. Each of the plurality of cooling units may be arranged at a position corresponding to each of the plurality of processing chips.

[0529] For example, when the number of processing chips used varies according to the control conditions of autonomous driving, cooling is performed using a cooling unit corresponding to the processing chips used, thereby enabling efficient cooling.

[0530] Fig.46 An example of the SoCBox 400 and the cooling unit 600 is schematically shown. Fig.46 The cooling unit 600 is illustrated as being composed of one cooling unit. When the cooling execution device 500 predicts that the SoCBox 400 starts to generate heat or that the temperature of the SoCBox 400 exceeds a preset threshold, the cooling unit 600 starts cooling, thereby cooling the entire SoCBox 400.

[0531] Fig.47 An example of the SoCBox 400 and the cooling unit 600 is schematically shown. Fig.47 The example shows a case where the cooling unit 600 is composed of a plurality of cooling units for cooling a plurality of parts of the SoCBox 400. The cooling execution device 500 predicts the temperature change of each of the plurality of parts of the SoCBox 400, and in response to the prediction that any part starts to generate heat or the temperature of any part will exceed a preset threshold, performs cooling using only the cooling unit corresponding to the part, thereby achieving efficient cooling.

[0532] Fig.48 An example of the SoCBox 400 and the cooling unit 600 is schematically shown. Fig.48The case where the cooling unit 600 is composed of two types of cooling units is illustrated. The cooling execution device 500 predicts the temperature change of each of the multiple parts of the SoCBox400, and in response to the prediction that any part begins to heat up or the temperature of any part will exceed a preset threshold, performs cooling using only the cooling unit corresponding to the part, thereby achieving efficient cooling. In addition, the cooling execution device 500 increases the cooling units used as the temperature of the SoCBox400 increases, that is, in this example, first, one of the two cooling units is used for cooling, and when the temperature of the SoCBox400 further increases, the other cooling unit is also used for cooling, thereby making the energy used for cooling more efficient.

[0533] <Ninth embodiment>

[0534] When a SoC (System on Chip) (processing chip) for autonomous driving performs advanced computing, heat generation becomes a technical problem. Therefore, in this embodiment, a technology for cooling a SoCBox as a control device by flexibly applying AI (Artificial Intelligence) is provided.

[0535] The SoCBox will immediately become hot, making it difficult to perform advanced calculations in the vehicle (a technical issue for fully autonomous driving). For example, the SoCBox has multiple SoCs, and as the control of autonomous driving becomes more complex, the processing performed by the SoC increases, so there is a possibility that the SoCBox will become hot and overheat. Therefore, AI predicts the heat dissipation of the SoCBox, communicates with other vehicles traveling near the vehicle before the SoCBox overheats, and allows the other vehicles to take over part of the computing load of the SoC, thereby preventing the SoCBox from becoming hot and making advanced calculations in the vehicle possible.

[0536] Fig.49 An example of the system 10 is schematically shown. The system 10 includes a management server 100. The system 10 includes a SoCBox 400. The system 10 includes a cooling device 700. The system 10 includes a cooling unit 600.

[0537] SoCBox400, cooling device 700 and cooling unit 600 are mounted on the vehicle. SoCBox400 is a control device having multiple SoCs, and uses the sensor values ​​of multiple sensors mounted on the vehicle to control the automatic driving of the vehicle. The automatic driving control of the vehicle consumes a very high processing load, so there is a situation where SoCBox400 becomes very hot. If the SoCBox400 is too hot, it is possible that the operation of SoCBox400 cannot be performed normally, or it may have a negative impact on the vehicle.

[0538] The cooling device 700 involved in this embodiment predicts the temperature change of SoCBox400, and communicates with other vehicles existing in a given range based on the temperature change, and instructs the control device to perform operations related to the control of the automatic driving of other vehicles that are communicating. For example, the cooling device 700 predicts that the SoCBox400 begins to heat up, and when the predicted temperature after heating will exceed the temperature of a preset threshold, it communicates with other vehicles traveling side by side near the vehicle. Thereafter, the cooling device 700, for example, groups multiple vehicles that are communicating, and instructs the SoCBox400 to perform operations to take over part of the computing load of the SoC of the vehicle that is the communication source. In addition, the cooling device 700 can also, for example, complement operations related to automatic driving between communicating vehicles.

[0539] Thus, when it is predicted that the temperature of SoCBox400 will exceed the set threshold, the cooling device 700 communicates with nearby vehicles to take over part of the computing load, or complement each other's computing load, thereby reducing the computing load and thus lowering the temperature of SoCBox400.

[0540] Here, the aforementioned preset threshold temperature refers to a temperature set as an upper limit temperature at which a plurality of SoCs included in the SoCBox 400 can normally perform processing.

[0541] In addition, the cooling device 700 can also communicate with other vehicles that exist within a range that allows communication with other vehicles to be maintained for a given time by judging them as vehicles that exist within a given range. For example, the communication unit of the cooling device 700 can communicate, and judges a vehicle that exists within a range that allows communication between vehicles to be maintained for a certain period of time as a vehicle that exists within a given range, and communicates with the vehicle. In addition, as a method of judging a vehicle that exists within a given range, for example, a vehicle that responds to a given broadcast communication, a vehicle that is confirmed to exist nearby based on location information, etc., is judged as a vehicle that exists within a given range.

[0542] In addition, the cooling device 700 may also start cooling the SoCBox based on the temperature change predicted by the prediction unit. For example, when it is predicted that the temperature of the SoCBox 400 will exceed the aforementioned threshold temperature, the cooling device 700 starts cooling the SoCBox 400 using an air cooling unit or the like.

[0543] Thus, the cooling device 700 cools the SoCBox 400 from the outside, while communicating with the other vehicles mentioned above to reduce the calculation load, so that the temperature of the SoCBox 400 can be rapidly cooled.

[0544] The cooling device 700 can predict the temperature change of the SoCBox 400 through AI. In addition, the learning of the temperature change of the SoCBox 400 can be performed using the data collected by the vehicle 200. For example, the management server 100 collects data from the vehicle 200 to perform learning. The subject that performs learning is not limited to the management server 100, and it can also be other devices.

[0545] The vehicle 200 is equipped with the SoCBox 400, the temperature sensor 40 for measuring the temperature of the SoCBox 400, and the cooling device 700. The SoCBox 400 controls the automatic driving of the vehicle 200 using the sensor values ​​of the multiple sensors mounted on the vehicle 200 and the external information received from the multiple servers 30.

[0546] The cooling device 700 obtains sensor values ​​of multiple sensors mounted on the vehicle 200 obtained by the SoCBox400, external information received from multiple servers 30, or from the SoCBox400, and inputs the obtained information into the learning model to predict the temperature change of the SoCBox400.

[0547] For example, when it is predicted that SoCBox400 starts to generate heat and the predicted temperature after generation of heat exceeds a preset threshold value, the cooling device 700 communicates with other vehicles traveling nearby to reduce the computational load of SoCBox400, thereby lowering the temperature of SoCBox400.

[0548] The server 30 may be an example of an external device. Examples of various servers 30 include a server that provides traffic information, a server that provides weather information, etc. The SoCBox 400 sends sensor values ​​used in the control of autonomous driving, external information, etc., and temperature changes of the SoCBox 400 during control to the management server 100.

[0549] The management server 100 performs learning using information received from one or more SoCBox400. The management server 100 generates a learning model that uses the information obtained by SoCBox400 as input and the temperature change of SoCBox400 as output by performing machine learning using information such as sensor values ​​obtained by SoCBox400, external information, and the temperature change of SoCBox400 when SoCBox400 obtains such information as learning data.

[0550] The vehicle 300 is a vehicle having a cooling unit according to the present embodiment. The SoCBox 400, the cooling device 700, and the cooling unit 600 are mounted on the vehicle 300. The cooling device 700 can receive the learning model generated by the management server 100 from the management server 100 and store the learning model.

[0551] The SoC Box 400, the cooling device 700, the management server 100, and the server 30 can communicate via the network 20. The network 20 can include a vehicle network. The network 20 can include the Internet. The network 20 can include a LAN (Local Area Network).

[0552] The network 20 may include a mobile communication network. The mobile communication network may comply with any of the 5G (5th Generation) communication method, the LTE (Long Term Evolution) communication method, the 3G (3rd Generation) communication method, and the 6G (6th Generation) communication method or later communication methods.

[0553] Fig.50 This is an explanatory diagram of the state of inter-vehicle communication in this embodiment. As described above, when it is predicted that the temperature of SoCBox400 will become a temperature exceeding a preset threshold, the cooling device 700 communicates with other vehicles traveling nearby, and the vehicle serving as the communication target instructs SoCBox400 to perform operations to take over part of the computational load of the SoC of the communication source vehicle.

[0554] exist Fig.50 In the example, vehicles 200-2, 200-3, and 200-4 are traveling at the same speed near vehicle 200-1. Then, it is predicted that the temperature of SoCBox400 of vehicle 200-1 will exceed the threshold temperature, so based on the communication unit of 200-1, vehicles 200-1 to 200-4 are grouped to form network 20.

[0555] After forming the network 20, the cooling device 700 of the communication target vehicle instructs to perform operations to take over part of the computational load of the SoCBox 400 of the communication source vehicle. As a result, part of the computational load of the SoC of the communication source vehicle 200-1 is taken over by the SoCs of the other communication target vehicles 200-2 to 200-4, thereby reducing the computational load of the SoC of the communication source vehicle 200-1 and lowering the temperature of the SoCBox 400 of the vehicle 200-1.

[0556] Fig.512 is an explanatory diagram for explaining the learning stage in the system 10. Here, as sensors 210 mounted on the vehicle 200, a camera 211, a LiDAR (Light Detection And Ranging) 212, a millimeter wave sensor 213, an ultrasonic sensor 214, an IMU sensor 215, and a GNSS (Global Navigation Satellite System) sensor 216 are exemplified. The vehicle 200 does not necessarily have to be equipped with all of these sensors, and may not have some of them, or may have sensors other than these sensors.

[0557] SoCBox400 acquires sensor information from each sensor included in the sensor 210. In addition, SoCBox400 can perform communication via the network 20, and SoCBox400 receives external information from multiple servers 30 via the network 20. Then, SoCBox400 performs automatic driving control of the vehicle 200 and the vehicle 300 using the acquired information.

[0558] The temperature sensor 40 measures the temperature change of the SoCBox 400. The SoCBox 400 transmits the sensor information received from the sensor 210, the external information received from the server 30, and the temperature change measured by the temperature sensor 40 when the automatic driving control is performed by acquiring these information to the management server 100.

[0559] The management server 100 includes an information acquisition unit 102 , a model generation unit 104 , and a model provision unit 106 . The information acquisition unit 102 acquires various information. The management server 100 can receive information transmitted from the SoCBox 400 .

[0560] The model generation unit 104 generates a learning model by performing machine learning using the information acquired by the information acquisition unit 102. The model generation unit 104 generates a learning model that uses the information acquired by SoCBox400 as input and the temperature change of SoCBox400 as output by performing machine learning using the information acquired by SoCBox400 and the temperature change of SoCBox400 when SoCBox400 acquires the information as learning data.

[0561] The model providing unit 106 provides the learning model generated by the model generating unit 104. The model providing unit 106 may transmit the learning model to the cooling device 700 mounted on the vehicle 200 and the vehicle 300.

[0562] The system 10 may also be configured to predict the temperature change of each of the multiple parts of the SoCBox400. In this case, the vehicle 200 may be provided with multiple temperature sensors 40 for measuring the temperature change of each of the multiple parts of the SoCBox400. The SoCBox400 may send the sensor information received from the sensor 210, the external information received from the server 30, and the temperature change measured by the multiple temperature sensors 40 when the automatic driving control is performed by acquiring these information to the management server 100.

[0563] The model generation unit 104 performs machine learning by using the information acquired by SoCBox400 and the temperature changes of multiple parts of SoCBox400 when SoCBox400 acquires the information as learning data, thereby generating a learning model that uses the information acquired by SoCBox400 as input and the temperature changes of multiple parts of SoCBox400 as output.

[0564] Fig.52 This is an explanatory diagram for explaining the cooling execution stage in the system 10. Here, as sensors 310 mounted on the vehicle 300, a camera 311, a LiDAR 312, a millimeter wave sensor 313, an ultrasonic sensor 314, an IMU sensor 315, and a GNSS sensor 316 are exemplified. The vehicle 300 does not necessarily have to be equipped with all of these sensors, and may not have some of them, or may have sensors other than these sensors.

[0565] The cooling device 700 includes a prediction unit 506, a communication unit 520, an instruction unit 530, a cooling execution unit 540, a model storage unit 550, and an information acquisition unit 560. The prediction unit 506 predicts the temperature change of the SoCBox400. The prediction unit 506 can predict the temperature change of the SoCBox400 through AI. For example, the prediction unit 506 predicts the temperature change of the SoCBox400 by inputting the information acquired by the information acquisition unit 560 described later into the learning model stored in the model storage unit 550 described later.

[0566] The communication unit 520 communicates with other vehicles existing within a given range based on the temperature change predicted by the prediction unit 506. For example, when the prediction unit 506 predicts that the temperature of the SoCBox 400 will become a temperature exceeding a preset threshold, the communication unit 520 communicates with other vehicles determined to exist within the aforementioned given range to form the network 20.

[0567] The instruction unit 530 instructs the control device to perform operations related to the control of the automatic driving of other vehicles that the communication unit 520 communicates with. For example, when the own vehicle is connected to another vehicle for communication, the instruction unit 530 refers to the idleness of the resources of the SoC of the own vehicle to determine whether it is possible to perform operations on a part of the computational load of the SoC of the other vehicle of the communication source. Then, for example, when it is determined that the operation of a part of the computational load of the SoC of the other vehicle of the communication source can be performed, the instruction unit 530 instructs the SoCBox 400 of the own vehicle to perform operations on a part of the computational load of the SoC of the other vehicle of the communication source.

[0568] The cooling execution unit 540 starts cooling the SoCBox 400 based on the temperature change of the SoCBox 400 predicted by the prediction unit 506. For example, the cooling execution unit 540 starts cooling the SoCBox 400 in response to the prediction unit 506 predicting that the temperature of the SoCBox 400 will become higher than a preset threshold.

[0569] The cooling execution unit 540 may use the cooling unit 600 described later to execute cooling of the SoCBox 400. The cooling unit 600 may use an air cooling unit to cool the SoCBox 400. The cooling unit 600 may use a water cooling unit to cool the SoCBox 400. The cooling unit 600 may use a liquid nitrogen cooling unit to cool the SoCBox 400.

[0570] The model storage unit 550 stores the learning model received from the management server 100. The information acquisition unit 504 acquires the information acquired by the SoCBox 400.

[0571] The information acquisition unit 560 acquires the sensor information acquired by the SoCBox 400 from the sensor 310 from the sensor 310 or the SoCBox 400. For example, the information acquisition unit 560 may receive the sensor information acquired by the SoCBox 400 from the sensor 310 from the SoCBox 400.

[0572] The information acquisition section 560 may also receive the same sensor information as the sensor information acquired by the SoCBox 400 from the sensor 310 from the sensor 310. In this case, each of the sensors 310 may transmit the sensor information to each of the SoCBox 400 and the cooling device 700.

[0573] The information acquisition unit 560 acquires the external information acquired by the SoCBox400 from the server 30 from the server 30 or the SoCBox400. The information acquisition unit 560 may receive the external information received by the SoCBox400 from the server 30 from the SoCBox400. The information acquisition unit 560 may also receive the same external information from the server 30 as the external information received by the SoCBox400 from the server 30. In this case, the server 30 may send the external information to each of the SoCBox400 and the cooling device 700.

[0574] The cooling unit 600 may also include multiple cooling units. For example, the cooling unit 600 includes multiple air cooling units. For example, the cooling unit 600 includes multiple water cooling units. For example, the cooling unit 600 includes multiple liquid nitrogen cooling units. The cooling unit 600 may also include multiple of one or more air cooling units, one or more water cooling units, and one or more liquid nitrogen cooling units.

[0575] The plurality of cooling units may be configured to cool different parts of the SoCBox 400. The prediction unit 506 may use the information acquired by the information acquisition unit 504 to predict the temperature change of each of the plurality of parts of the SoCBox 400. The cooling execution unit 508 may start cooling the SoCBox 400 using a cooling unit selected from the plurality of cooling units that cool the plurality of parts of the SoCBox 400, based on the prediction result of the prediction unit 506.

[0576] The cooling execution unit 540 may cool the SoC Box 400 using cooling units corresponding to the temperature of the SoC Box 400 predicted by the prediction unit 506. For example, the higher the temperature of the SoC Box 400, the more cooling units the cooling execution unit 508 uses to cool the SoC Box 400.

[0577] As a specific example, when it is predicted that the temperature of SoCBox400 will exceed a first threshold, the cooling execution unit 540 starts cooling using one of a plurality of cooling units. Even so, if the temperature of SoCBox400 still rises and is predicted to exceed a second threshold, the number of cooling units used is increased.

[0578] The cooling execution unit 508 can use a more powerful cooling unit to cool the SoCBox400 when the temperature of the SoCBox400 is higher. For example, when it is predicted that the temperature of the SoCBox400 will exceed the first threshold, the cooling execution unit 508 starts cooling using the air cooling unit. When the temperature of the SoCBox400 still rises despite this and is predicted to exceed the second threshold, the cooling execution unit 508 starts cooling using the water cooling unit. When the temperature of the SoCBox400 still rises despite this and is predicted to exceed the third threshold, the cooling execution unit 508 starts cooling using the liquid nitrogen cooling unit.

[0579] The SoCBox 400 may have a plurality of processing chips, and the plurality of processing chips may be respectively arranged at different positions of the SoCBox 400. Each of the plurality of cooling units may be arranged at a position corresponding to each of the plurality of processing chips.

[0580] For example, when the number of processing chips used varies according to the control conditions of autonomous driving, cooling is performed using a cooling unit corresponding to the processing chips used, thereby enabling efficient cooling.

[0581] Fig.53 An example of the SoCBox 400 and the cooling unit 600 is schematically shown. Fig.53 The cooling unit 600 is illustrated as a case where it is composed of one cooling unit. When it is predicted that the SoCBox 400 starts to generate heat or that the temperature of the SoCBox 400 exceeds a preset threshold, the cooling device 700 starts cooling using the cooling unit 600 to cool the entire SoCBox 400.

[0582] Fig.54 An example of the SoCBox 400 and the cooling unit 600 is schematically shown. Fig.54 The case where the cooling unit 600 is composed of a plurality of cooling units that cool a plurality of locations of the SoCBox 400 , respectively, is illustrated.

[0583] The cooling device 700 predicts the temperature changes of multiple parts of the SoCBox400, and in response to the prediction that any part starts to heat up or the temperature of any part will exceed a preset threshold, performs cooling using only the cooling unit corresponding to the part, thereby achieving efficient cooling.

[0584] Fig.55 An example of the SoCBox 400 and the cooling unit 600 is schematically shown. Fig.55The case where the cooling unit 600 is composed of two types of cooling units is illustrated. The cooling device 700 predicts the temperature change of each of the multiple parts of the SoCBox 400, and in response to the prediction that any part starts to heat up or the temperature of any part will exceed a preset threshold, performs cooling using only the cooling unit corresponding to the part, thereby achieving efficient cooling.

[0585] In addition, the cooling device 700 increases the number of cooling units used as the temperature of the SoCBox 400 increases. That is, in this example, first, one of the two cooling units is used for cooling, and when the temperature of the SoCBox 400 further increases, the other cooling unit is also used for cooling, thereby making the energy used for cooling more efficient.

[0586] <Tenth embodiment>

[0587] When the SoC (System on Chip) for autonomous driving performs advanced computing, heat generation becomes a technical problem. Therefore, in this embodiment, a technology called Synchronized Burst Chilling is provided that flexibly applies AI (Artificial Intelligence) to optimize the rapid cooling of the SoCBox.

[0588] The SoCBox will immediately become hot, making it difficult to perform advanced computing in the vehicle (a technical issue for fully autonomous driving). For this reason, it is considered to predict the heat dissipation of the SoCBox through AI and cool the SoCBox. For example, by predicting the heat dissipation of the SoCBox through AI and cooling it at the same time as the heat dissipation, the SoCBox can be prevented from becoming hot, making advanced computing in the vehicle possible.

[0589] In this case, it is desirable to optimize the timing of cooling. For example, the moment when the computing power of the SoCBox reaches the maximum is predicted, and the best time is estimated to control the cooling device. In addition, for example, the time from the triggering of cooling to the actual cooling is calculated backwards and taken into consideration. In addition, for example, the degree of cooling is adjusted according to the risk rate. As a specific example, the following control is performed: if the risk rate exceeds 30%, cooling is performed; if the risk rate is below 20%, cooling is weakened; if the risk rate exceeds 50%, cooling is strengthened.

[0590] AI predicts the heat dissipation of the SoCBox and performs cooling simultaneously with the heat dissipation, thereby preventing the SoCBox from becoming hot, making advanced computing in the vehicle possible. In addition, cooling can be performed efficiently considering necessity, time, and risk.

[0591] Fig.56An example of the system 10 is schematically shown. The system 10 includes a management server 100 . The system 10 includes a SoCBox 400 . The system 10 includes a cooling execution device 500 . The system 10 includes a cooling unit 600 .

[0592] SoCBox400, cooling execution device 500 and cooling unit 600 are mounted on the vehicle. SoCBox400 uses the sensor values ​​of multiple sensors mounted on the vehicle to control the automatic driving of the vehicle. The automatic driving control of the vehicle consumes a very high processing load, so there is a situation where SoCBox400 becomes very hot. If the SoCBox400 is too hot, it is possible that the operation of SoCBox400 cannot be performed normally, or it may have a bad effect on the vehicle.

[0593] The cooling execution device 500 involved in this embodiment predicts the temperature change of SoCBox400, and starts cooling SoCBox400 based on the temperature change. For example, the cooling execution device 500 responds to the prediction that SoCBox400 starts to heat up and immediately starts cooling SoCBox400. By starting cooling earlier than the start of heating or at the same time as the start of heating, it is possible to reliably prevent SoCBox400 from becoming high temperature. In addition, compared with the case where SoCBox400 is always cooled, the energy required for cooling can be reduced.

[0594] The cooling execution device 500 can predict the temperature change of the SoCBox 400 through AI. The learning of the temperature change of the SoCBox 400 can be performed using the data collected by the vehicle 200. For example, the management server 100 collects data from the vehicle 200 to perform learning. The subject that performs learning is not limited to the management server 100, and it can also be other devices.

[0595] The vehicle 200 is equipped with a SoCBox 400 and a temperature sensor 40 for measuring the temperature of the SoCBox 400. The SoCBox 400 controls the automatic driving of the vehicle 200 using the sensor values ​​of the multiple sensors mounted on the vehicle 200 and the external information received from the various servers 30. The server 30 may be an example of an external device. As examples of the various servers 30, a server that provides traffic information, a server that provides weather information, etc. may be cited. The SoCBox 400 sends the sensor values, external information, etc. used in the control of the automatic driving, and the temperature change of the SoCBox 400 during the control to the management server 100.

[0596] The management server 100 performs learning using information received from one or more SoCBox400. The management server 100 generates a learning model that uses the information obtained by SoCBox400 as input and the temperature change of SoCBox400 as output by performing machine learning using information such as sensor values ​​obtained by SoCBox400, external information, and the temperature change of SoCBox400 when SoCBox400 obtains such information as learning data.

[0597] The vehicle 300 is a vehicle having a cooling function according to the present embodiment. The vehicle 300 is equipped with a SoCBox 400, a cooling execution device 500, and a cooling unit 600. The cooling execution device 500 can receive a learning model generated by the management server 100 from the management server 100 and store the learning model.

[0598] The cooling execution device 500 can obtain the sensor values ​​of multiple sensors mounted on the vehicle 300 obtained by the SoCBox400, the external information received from multiple servers 30 from multiple sensors, multiple servers 30, or from the SoCBox400, and input the obtained information into the learning model to predict the temperature change of the SoCBox400.

[0599] The cooling execution device 500 can start cooling the SoC Box 400 using the cooling unit 600 when it is predicted that the SoC Box 400 starts to generate heat and the temperature of the SoC Box 400 will become higher than a preset threshold.

[0600] The cooling execution device 500 can also predict the computing power of SoCBox400 through the calculation process when predicting the temperature change of SoCBox400. For example, the cooling execution device 500 predicts the change of the computing power of SoCBox400 through AI, and predicts the temperature change of SoCBox400 based on the prediction result. As a specific example, the cooling execution device 500 predicts the moment when the computing power of SoCBox400 reaches the maximum through AI, and adds the temperature change of SoCBox400 caused by the computing power of SoCBox400 reaching the maximum to predict the temperature change of SoCBox400.

[0601] The management server 100 can generate a learning model that uses the information obtained by SoCBox400 as input and the computing power of SoCBox400 as output by executing machine learning using information such as sensor values, external information, etc. obtained by SoCBox400 and the measured computing power of SoCBox400 as learning data. The cooling execution device 500 can use the learning model to predict changes in the computing power of SoCBox400.

[0602] In addition, the cooling execution device 500 involved in this embodiment can also predict the change of the computing power of SoCBox400, and start cooling the SoCBox400 based on the change of the computing power. For example, the cooling execution device 500 predicts the moment when the computing power of SoCBox400 reaches the maximum, and starts cooling the SoCBox400 based on the prediction result. As a specific example, the cooling execution device 500 predicts the moment when the computing power of SoCBox400 reaches the maximum, and starts cooling the SoCBox400 at the predicted moment. In addition, for example, the cooling execution device 500 predicts the moment when the computing power of SoCBox400 reaches the maximum, and starts cooling the SoCBox400 at a predetermined time back from the predicted moment.

[0603] The cooling execution device 500 involved in this embodiment can perform the cooling of SoCBox400 by taking into account the time from the triggering of cooling to the actual cooling. For example, the cooling execution device 500 uses AI to predict the time until SoCBox400 is actually cooled when the cooling of SoCBox400 is started in a state where SoCBox400 is set, and controls the timing of starting the cooling of SoCBox400 based on the prediction result. As a specific example, the cooling execution device 500 starts the cooling of SoCBox400 by tracing back the predicted time from the timing when the cooling of SoCBox400 is actually desired to be started. Even if the cooling is started with the timing when the cooling of SoCBox400 is actually desired to be started, if there is a time lag before SoCBox400 is actually cooled, the start of cooling will be delayed, but by using such a prediction, the cooling of SoCBox400 can be actually performed at the timing when the cooling of SoCBox400 is actually desired to be started.

[0604] The management server 100 can generate a learning model that uses the information obtained by SoCBox 400 as input and the time until SoCBox 400 is actually cooled when cooling of SoCBox 400 is started, by executing machine learning using information such as sensor values ​​obtained by SoCBox 400, external information, and the time until SoCBox 400 is cooled when cooling of SoCBox 400 is started when SoCBox 400 obtains the information as learning data. The cooling execution device 500 can control the triggering of cooling of SoCBox 400 using the learning model.

[0605] The cooling execution device 500 involved in this embodiment can adjust the degree of cooling of the SoCBox400 according to the risk rate, for example. The risk rate can be, for example, the probability that the SoCBox400 no longer works normally due to the high temperature of the SoCBox400. The risk rate can be, for example, the probability that the SoCBox400 no longer works normally due to the high temperature of the SoCBox400, and the normal autonomous driving of the vehicle controlled by the SoCBox400 is affected.

[0606] The risk rate may be, for example, the probability of an accident occurring due to the SoCBox400 becoming too hot and causing the SoCBox400 to no longer work properly, thereby affecting the normal autonomous driving of the vehicle controlled by the SoCBox400. For example, the risk rate is different when the SoCBox400 no longer works properly due to high temperature when the weather is good and the vehicle is traveling in an area with nothing around and the SoCBox400 no longer works properly due to high temperature and when the weather is bad and the vehicle is traveling in a congested area such as a city.

[0607] The cooling execution device 500, for example, pre-stores a first threshold value for the risk rate and a second threshold value higher than the first threshold value. Moreover, when the risk rate is lower than the first threshold value, the cooling execution device 500 cools the SoCBox400 with a first cooling intensity, and when the risk rate is higher than the first threshold value and lower than the second threshold value, the cooling execution device 500 cools the SoCBox400 with a second cooling intensity that is stronger than the first cooling intensity, and when the risk rate is higher than the second threshold value, the cooling execution device 500 cools the SoCBox400 with a third cooling intensity that is stronger than the second cooling intensity. Thus, by cooling the SoCBox400 according to the risk rate, it is possible to optimize the energy required for cooling. In addition, the stages of cooling intensity are not limited to 3 stages.

[0608] SoCBox400, cooling execution device 500, management server 100, and server 30 can communicate via network 20. Network 20 can include a vehicle network. Network 20 can include the Internet. Network 20 can include a LAN (Local Area Network). Network 20 can include a mobile communication network. The mobile communication network can comply with any of the communication methods after 5G (5th Generation) communication method, LTE (Long Term Evolution) communication method, 3G (3rd Generation) communication method, and 6G (6th Generation) communication method.

[0609] Fig.572 is an explanatory diagram for explaining the learning stage in the system 10. Here, as sensors 210 mounted on the vehicle 200, a camera 211, a LiDAR (Light Detection And Ranging) 212, a millimeter wave sensor 213, an ultrasonic sensor 214, an IMU sensor 215, and a GNSS (Global Navigation Satellite System) sensor 216 are exemplified. The vehicle 200 does not necessarily have to be equipped with all of these sensors, and may not have some of them, or may have sensors other than these sensors.

[0610] The SoCBox 400 acquires sensor information from each sensor included in the sensor 210. In addition, the SoCBox 400 can perform communication via the network 20, and the SoCBox 400 receives external information from a plurality of servers 30 respectively via the network 20. Then, the SoCBox 400 performs automatic driving control of the vehicle 200 using the acquired information.

[0611] The temperature sensor 40 measures the temperature change of the SoCBox 400. The SoCBox 400 transmits the sensor information received from the sensor 210, the external information received from the server 30, and the temperature change measured by the temperature sensor 40 when the automatic driving control is performed by acquiring these information to the management server 100.

[0612] SoCBox400 records the computing power of SoCBox400. SoCBox400 can record the computing power of SoCBox400 either regularly or irregularly. SoCBox400 can record and send to the management server 100 the sensor information received from the sensor 210, the external information received from the server 30, and the computing power of SoCBox400 when the information is obtained to perform the autonomous driving control.

[0613] The management server 100 includes an information acquisition unit 102 , a model generation unit 104 , and a model provision unit 106 . The information acquisition unit 102 acquires various information. The management server 100 can receive information transmitted from the SoCBox 400 .

[0614] The model generation unit 104 performs machine learning using the information acquired by the information acquisition unit 102 to generate a learning model.

[0615] The model generation unit 104 can generate a learning model having information acquired by SoCBox400 as input and temperature change of SoCBox400 as output by executing machine learning using information acquired by SoCBox400 and temperature change of SoCBox400 when SoCBox400 acquires the information as learning data.

[0616] The model generation unit 104 can generate a learning model with information acquired by SoCBox400 as input and the computing power of SoCBox400 as output by performing machine learning using the information acquired by SoCBox400 and the computing power of SoCBox400 when SoCBox400 acquired the information as learning data.

[0617] The model generation unit 104 can generate a learning model with the information obtained by SoCBox400 as input and the time from when the cooling of SoCBox400 is started to when SoCBox400 is cooled as output by performing machine learning with the information obtained by SoCBox400 and the time from when the cooling of SoCBox400 is started to when SoCBox400 obtains the information as learning data.

[0618] The model providing unit 106 provides the learning model generated by the model generating unit 104. The model providing unit 106 may transmit the learning model to the cooling execution device 500 mounted on the vehicle 300.

[0619] The system 10 may also be configured to predict the temperature change of each of the multiple parts of the SoCBox400. In this case, the vehicle 200 may be provided with multiple temperature sensors 40 that measure the temperature change of each of the multiple parts of the SoCBox400. The SoCBox400 may send the sensor information received from the sensor 210, the external information received from the server 30, and the temperature change measured by the multiple temperature sensors 40 when the automatic driving control is performed after obtaining the information to the management server 100. The model generation unit 104 performs machine learning by using the information obtained by the SoCBox400 and the temperature change of each of the multiple parts of the SoCBox400 when the SoCBox400 obtains the information as learning data, thereby generating a learning model that uses the information obtained by the SoCBox400 as input and the temperature change of each of the multiple parts of the SoCBox400 as output.

[0620] The cooling unit 600 may be mounted on the vehicle 200 to collect data related to the cooling of the SoCBox400. For example, when the cooling unit 600 is used to cool the SoCBox400, the time from the time when the cooling unit 600 is triggered to the time when the SoCBox400 is actually cooled is measured using the temperature sensor 40. The SoCBox400 sends the sensor information received from the sensor 210, the external information received from the server 30, and the time from the time when the cooling unit 600 is triggered to the time when the SoCBox400 is actually cooled to the management server 100 when the cooling unit 600 cools the SoCBox400 when the information is obtained. The model generation unit 104 of the management server 100 can generate a learning model that uses the information obtained by SoCBox400 as input and uses the time from when the cooling of SoCBox400 is started to when SoCBox400 is actually cooled as output by executing machine learning using information such as sensor values ​​obtained by SoCBox400, external information, and the time from when SoCBox400 obtains the information and starts cooling the SoCBox400 to when SoCBox400 is actually cooled as learning data. The model providing unit 106 can send the learning model to the cooling execution device 500 mounted on the vehicle 300.

[0621] Fig.58 This is an explanatory diagram for explaining the cooling execution stage in the system 10. Here, as sensors 310 mounted on the vehicle 300, a camera 311, a LiDAR 312, a millimeter wave sensor 313, an ultrasonic sensor 314, an IMU sensor 315, and a GNSS sensor 316 are exemplified. The vehicle 300 does not necessarily have to be equipped with all of these sensors, and may not have some of them, or may have sensors other than these sensors.

[0622] The cooling execution device 500 includes a model storage unit 502, an information acquisition unit 504, a prediction unit 506, a cooling execution unit 508, and a risk rate acquisition unit 511. The model storage unit 502 stores the learning model received from the management server 100. The information acquisition unit 504 acquires the information acquired by the SoCBox 400.

[0623] The information acquisition unit 504 acquires the sensor information acquired by the SoCBox400 from the sensor 310 from the sensor 310 or the SoCBox400. For example, the information acquisition unit 504 may receive the sensor information acquired by the SoCBox400 from the sensor 310 from the SoCBox400. The information acquisition unit 504 may also receive the same sensor information from the sensor 310 as the sensor information acquired by the SoCBox400 from the sensor 310. In this case, each of the sensors 310 may send the sensor information to the SoCBox400 and each of the cooling execution devices 500.

[0624] The information acquisition unit 504 acquires the external information acquired by the SoCBox400 from the server 30 from the server 30 or the SoCBox400. The information acquisition unit 504 may receive the external information received by the SoCBox400 from the server 30 from the SoCBox400. The information acquisition unit 504 may also receive the same external information from the server 30 as the external information received by the SoCBox400 from the server 30. In this case, the server 30 may send the external information to each of the SoCBox400 and the cooling execution device 500.

[0625] The prediction unit 506 can predict the temperature change of the SoCBox 400. The prediction unit 506 can predict the temperature change of the SoCBox 400 through AI. For example, the prediction unit 506 predicts the temperature change of the SoCBox 400 by inputting the information acquired by the information acquisition unit 504 into the learning model stored in the model storage unit 502.

[0626] The prediction unit 506 can also predict the computing power of SoCBox400 through the calculation process when predicting the temperature change of SoCBox400. For example, the prediction unit 506 predicts the computing power of SoCBox400 by inputting the information acquired by the information acquisition unit 504 into the learning model stored in the model storage unit 502, and predicts the temperature change of SoCBox400 based on the prediction result. As a specific example, the prediction unit 506 predicts the moment when the computing power of SoCBox400 reaches the maximum, and adds the temperature change of SoCBox400 caused by the computing power of SoCBox400 reaching the maximum, to predict the temperature change of SoCBox400.

[0627] The cooling execution unit 508 may start cooling the SoCBox 400 based on the temperature change of the SoCBox 400 predicted by the prediction unit 506. For example, the cooling execution unit 508 starts cooling the SoCBox 400 in response to the prediction unit 506 predicting that the SoCBox 400 starts to heat up. For example, the cooling execution unit 508 starts cooling the SoCBox 400 in response to the prediction unit 506 predicting that the temperature of the SoCBox 400 will become higher than a preset threshold.

[0628] The cooling execution unit 508 may use the cooling unit 600 to execute cooling of the SoCBox 400. The cooling unit 600 may use an air cooling unit to cool the SoCBox 400. The cooling unit 600 may use a water cooling unit to cool the SoCBox 400. The cooling unit 600 may use a liquid nitrogen cooling unit to cool the SoCBox 400.

[0629] The cooling unit 600 may also include multiple cooling units. For example, the cooling unit 600 includes multiple air cooling units. For example, the cooling unit 600 includes multiple water cooling units. For example, the cooling unit 600 includes multiple liquid nitrogen cooling units. The cooling unit 600 may also include multiple of one or more air cooling units, one or more water cooling units, and one or more liquid nitrogen cooling units.

[0630] The plurality of cooling units may be configured to cool different parts of the SoCBox 400. The prediction unit 506 may use the information acquired by the information acquisition unit 504 to predict the temperature change of each of the plurality of parts of the SoCBox 400. The cooling execution unit 508 may start cooling the SoCBox 400 using a cooling unit selected from the plurality of cooling units that cool the plurality of parts of the SoCBox 400, based on the prediction result of the prediction unit 506.

[0631] The cooling execution unit 508 can use a cooling unit corresponding to the temperature of the SoCBox400 predicted by the prediction unit 506 to execute cooling of the SoCBox400. For example, the higher the temperature of the SoCBox400, the more cooling units the cooling execution unit 508 uses to execute cooling of the SoCBox400. As a specific example, when it is predicted that the temperature of the SoCBox400 will exceed a first threshold, the cooling execution unit 508 uses one of the plurality of cooling units to start cooling. Even so, when the temperature of the SoCBox400 still rises and is predicted to exceed a second threshold, the number of cooling units used is increased.

[0632] The cooling execution unit 508 can use a more powerful cooling unit to cool the SoCBox400 when the temperature of the SoCBox400 is higher. For example, when it is predicted that the temperature of the SoCBox400 will exceed the first threshold, the cooling execution unit 508 starts cooling using the air cooling unit. When the temperature of the SoCBox400 still rises despite this and is predicted to exceed the second threshold, the cooling execution unit 508 starts cooling using the water cooling unit. When the temperature of the SoCBox400 still rises despite this and is predicted to exceed the third threshold, the cooling execution unit 508 starts cooling using the liquid nitrogen cooling unit.

[0633] The SoCBox 400 may have a plurality of processing chips, and the plurality of processing chips may be respectively arranged at different positions of the SoCBox 400. Each of the plurality of cooling units may be arranged at a position corresponding to each of the plurality of processing chips.

[0634] For example, when the number of processing chips used varies according to the control conditions of autonomous driving, cooling is performed using a cooling unit corresponding to the processing chips used, thereby enabling efficient cooling.

[0635] The prediction unit 506 can predict the change of the computing power of SoCBox400. The prediction unit 506 can predict the change of the computing power of SoCBox400 through AI. For example, the prediction unit 506 predicts the computing power of SoCBox400 by inputting the information acquired by the information acquisition unit 504 into the learning model stored in the model storage unit 502. As a specific example, the prediction unit 506 can predict the moment when the computing power of SoCBox400 reaches the maximum.

[0636] The cooling execution unit 508 can start cooling the SoCBox400 based on the change in the computing power of the SoCBox400 predicted by the prediction unit 506. For example, the cooling execution unit 508 starts cooling the SoCBox400 based on the moment when the computing power of the SoCBox400 reaches the maximum predicted by the prediction unit 506. As a specific example, the cooling execution unit 508 starts cooling the SoCBox400 at the moment when the computing power of the SoCBox400 reaches the maximum. In this way, the cooling of the SoCBox400 can be started when the computing power of the SoCBox400 reaches the maximum and the heat generation reaches the maximum, and the cooling can be performed efficiently.

[0637] In addition, as a specific example, the cooling execution unit 508 starts cooling the SoCBox400 in response to the state where the computing power of the SoCBox400 is higher than the preset threshold value for a preset time. Even if the computing power of the SoCBox400 is high, if it returns to the low computing power state immediately afterwards, the temperature of the SoCBox400 will not become too high, but if the computing power of the SoCBox400 continues to be high to a certain extent, the temperature of the SoCBox400 may continue to rise. According to the cooling execution unit 508, such a temperature rise can be prevented.

[0638] The prediction unit 506 can also predict the time until the temperature of SoCBox400 becomes higher than the preset temperature based on the prediction result of the moment when the computing power of SoCBox400 reaches the maximum. The cooling execution unit 508 can start cooling the SoCBox400 based on the time predicted by the prediction unit 506. For example, the cooling execution unit 508 counts back from the moment when the temperature of SoCBox400 will become higher than the preset temperature to determine the timing to start cooling the SoCBox400 so that the temperature of SoCBox400 will not become higher than the preset temperature, and starts cooling the SoCBox400 at the determined timing.

[0639] The cooling execution unit 508 can execute the cooling of the SoCBox400 by taking into account the time from the triggering of cooling to the actual cooling. For example, the cooling execution unit 508 uses AI to predict the time until the SoCBox400 is actually cooled when the cooling of the SoCBox400 is started in a state where the SoCBox400 is set, and controls the timing of starting the cooling of the SoCBox400 according to the prediction result. The cooling execution unit 508 can predict the time by inputting the information obtained by the information acquisition unit 504 into the learning model stored in the model storage unit 502, and the learning model takes the information obtained by the SoCBox400 as input and the time until the SoCBox400 is actually cooled when the cooling of the SoCBox400 is started as output. As a specific example, the cooling execution unit 508 instructs the cooling unit 600 to start cooling when the timing of actually starting the cooling of the SoCBox400 is traced back by the predicted time.

[0640] The risk rate acquisition unit 511 acquires the risk rate. The risk rate acquisition unit 511 can calculate the risk rate using the information acquired by the information acquisition unit 504. The risk rate acquisition unit 511 calculates the possibility of an accident based on the temperature of the SoCBox400 predicted by the prediction unit 506, for example, using relationship data indicating the relationship between the temperature of the SoCBox400 and the occurrence of an accident of a vehicle equipped with the SoCBox400, thereby calculating the risk rate.

[0641] The risk rate acquisition unit 511 calculates the risk rate based on the position of the vehicle 300, for example. The risk rate acquisition unit 511 calculates the risk rate in such a manner that the greater the number of objects existing around the vehicle 300, the higher the risk rate. For example, the risk rate acquisition unit 511 calculates the risk rate in such a manner that the greater the number of other vehicles existing around the vehicle 300, the higher the risk rate. For example, the risk rate acquisition unit 511 calculates the risk rate in such a manner that the greater the number of people existing around the vehicle 300, the higher the risk rate. The risk rate acquisition unit 511 can determine the number and quantity of objects existing around the vehicle 300 using information obtained from the camera 311 and LiDAR 312, etc.

[0642] The risk rate acquisition unit 511 calculates the risk rate based on, for example, the weather in the area where the vehicle 300 is located. The risk rate acquisition unit 511 calculates the risk rate in such a manner that the risk rate is higher in a so-called bad weather state than in a so-called good weather state. As a specific example, the risk rate acquisition unit 511 calculates the risk rate in such a manner that the risk rate is higher in a rainy day than in a sunny day. The risk rate acquisition unit 511 can determine the weather in the area where the vehicle 300 is located based on information received from a server that provides weather information. The risk rate acquisition unit 511 can also determine the weather in the area where the vehicle 300 is located by detecting the surroundings of the vehicle 300 and clouds away from the vehicle 300 through the camera 311.

[0643] The risk ratio acquisition unit 511 may calculate the risk ratio by using a determination method other than these methods. In addition, the risk ratio acquisition unit 511 may calculate the risk ratio by combining a plurality of determination methods.

[0644] The cooling execution unit 508 can adjust the degree of cooling of the SoCBox400 according to the risk rate acquired by the risk rate acquisition unit 511. The cooling execution unit 508, for example, pre-stores a first threshold value for the risk rate and a second threshold value higher than the first threshold value. Then, when the risk rate is lower than the first threshold value, the cooling execution unit 508 cools the SoCBox400 with a first cooling intensity, and when the risk rate is higher than the first threshold value and lower than the second threshold value, the cooling execution unit 508 cools the SoCBox400 with a second cooling intensity that is stronger than the first cooling intensity, and when the risk rate is higher than the second threshold value, the cooling execution unit 508 cools the SoCBox400 with a third cooling intensity that is stronger than the second cooling intensity. In addition, the stages of cooling intensity are not limited to 3 stages.

[0645] Fig.59 An example of the SoCBox 400 and the cooling unit 600 is schematically shown. Fig.59 The cooling unit 600 is illustrated as being composed of one cooling unit. When the cooling execution device 500 predicts that the SoCBox 400 starts to generate heat or that the temperature of the SoCBox 400 exceeds a preset threshold, the cooling unit 600 starts cooling, thereby cooling the entire SoCBox 400.

[0646] Fig.60 An example of the SoCBox 400 and the cooling unit 600 is schematically shown. Fig.60 The example shows a case where the cooling unit 600 is composed of a plurality of cooling units for cooling a plurality of parts of the SoCBox 400. The cooling execution device 500 predicts the temperature change of each of the plurality of parts of the SoCBox 400, and in response to the prediction that any part starts to generate heat or the temperature of any part will exceed a preset threshold, performs cooling using only the cooling unit corresponding to the part, thereby achieving efficient cooling.

[0647] Fig.61 An example of the SoCBox 400 and the cooling unit 600 is schematically shown. Fig.61The case where the cooling unit 600 is composed of two types of cooling units is illustrated. The cooling execution device 500 predicts the temperature change of each of the multiple parts of the SoCBox400, and in response to the prediction that any part begins to heat up or the temperature of any part will exceed a preset threshold, performs cooling using only the cooling unit corresponding to the part, thereby achieving efficient cooling. In addition, the cooling execution device 500 increases the cooling units used as the temperature of the SoCBox400 increases, that is, in this example, first, one of the two cooling units is used for cooling, and when the temperature of the SoCBox400 further increases, the other cooling unit is also used for cooling, thereby making the energy used for cooling more efficient.

[0648] <Hardware Configuration Diagram>

[0649] Fig.62 An example of the hardware configuration of a computer 1200 that functions as a management server 100, SoCBox 400, cooling execution device 500, or cooling device 700 is schematically shown. The program installed in the computer 1200 enables the computer 1200 to function as one or more "parts" of the device involved in this embodiment, or enables the computer 1200 to perform operations associated with the device involved in this embodiment or the one or more "parts", and / or enables the computer 1200 to perform the process involved in this embodiment or the stage of the process. Such a program can be executed by the CPU 1212 in order to enable the computer 1200 to perform specific operations associated with some or all of the blo...

Claims

1. A cooling execution device, comprising: a prediction unit that predicts a temperature change of a control device that controls automatic driving of a vehicle and is mounted on the vehicle; and A cooling execution unit starts cooling the control device based on the temperature change predicted by the prediction unit.

2. The cooling actuator according to claim 1, in, The cooling execution unit starts cooling the control device in response to the prediction by the prediction unit that the temperature of the control device will become higher than a preset threshold value.

3. The cooling actuator according to claim 1, in, The prediction unit predicts the temperature change of the control device through artificial intelligence (AI).

4. The cooling execution device according to claim 3, in, The cooling execution device also includes: a model storage unit storing a learning model generated by machine learning, which has the information acquired by the control device as an input and has a temperature change of the control device as an output, wherein the machine learning uses the information acquired by the control device and the temperature change of the control device when the control device acquires the information as learning data; as well as an information acquisition unit, which acquires the information acquired by the control device, The prediction unit predicts the temperature change of the control device by inputting the information acquired by the information acquisition unit into the learning model.

5. The cooling execution device according to claim 4, in, The information acquisition unit acquires sensor information, which is acquired by the control device from a sensor mounted on the vehicle, from the sensor or the control device.

6. The cooling execution device according to claim 4, in, The information acquisition unit acquires, from the control device, an analysis result of a captured image analyzed by the control device, the captured image being captured by a camera mounted on the vehicle.

7. The cooling execution device according to claim 4, in, The information acquisition unit acquires external information received by the control device from an external device from the external device or the control device.

8. The cooling execution device according to claim 7, in, The information acquisition unit acquires the traffic information of the road on which the vehicle is located, which is received by the control device from the external device, from the external device or the control device.

9. The cooling actuator according to claim 1, in, The cooling execution unit starts cooling the control device using one or more cooling units selected from a plurality of cooling units, depending on the level of the temperature of the control device predicted by the prediction unit.

10. The cooling execution device according to claim 9, in, The multiple cooling units include one or more air cooling units, one or more water cooling units, and multiple liquid nitrogen cooling units.

11. The cooling actuator according to claim 10, in, The prediction unit predicts temperature changes of each of the plurality of parts of the control device. The cooling execution unit starts cooling the control device using a cooling unit selected from a plurality of cooling units that cool a plurality of locations of the control device, based on the prediction result of the prediction unit.

12. The cooling actuator according to claim 11, in, The control device has a plurality of processing chips respectively arranged at different positions of the control device. Each of the plurality of cooling units is disposed at a position corresponding to each of the plurality of processing chips. 13 . A program for causing a computer to function as the cooling execution device according to claim 1 .

14. A cooling execution method, executed by a computer, wherein the cooling execution method include: In a prediction stage, a temperature change of a control device is predicted, the control device controls the automatic driving of the vehicle and is mounted on the vehicle; as well as In the cooling execution phase, cooling of the control device is started based on the temperature change predicted in the prediction phase.

15. A cooling system comprising a server and a cooling execution device, It is characterized in that The server has: an information acquisition unit that acquires, from the cooling execution device, position information of a vehicle having the cooling execution device and temperature information of a control device of the vehicle; as well as a prediction unit that predicts a temperature change of the control device based on the position information of the vehicle having the cooling execution device and the temperature information of the control device of the vehicle acquired by the information acquisition unit, The cooling execution device comprises: an information acquisition unit that acquires, from the server, a prediction result of a temperature change of the control device predicted by the prediction unit of the server; as well as A cooling execution unit starts cooling of the control device based on the prediction result of the temperature change of the control device acquired by the information acquisition unit.

16. The cooling system according to claim 15, It is characterized in that The server further includes a creation unit, which creates a map representing the relationship between the position of the vehicle and the temperature of the control device based on the position information of the vehicle having the cooling execution device and the temperature information of the control device of the vehicle acquired by the information acquisition unit. The prediction unit predicts temperature changes of the control device based on position information of the vehicle having the cooling execution device acquired by the information acquisition unit, temperature information of the control device of the vehicle, and a mapping chart representing the relationship between the position of the vehicle and the temperature of the control device created by the creation unit.

17. The cooling system according to claim 16, It is characterized in that The information acquisition unit of the server acquires, from the cooling execution device, position information of a vehicle having the cooling execution device, temperature information of a control device of the vehicle, and information related to the type of the vehicle. The creation unit creates a map representing the relationship between the position of the vehicle and the temperature of the control device in the type of the vehicle based on the position information of the vehicle having the cooling execution device, the temperature information of the control device of the vehicle, and the information related to the type of the vehicle acquired by the information acquisition unit. The prediction unit of the server predicts the temperature change of the control device based on the position information of the vehicle having the cooling execution device acquired by the information acquisition unit, the temperature information of the control device possessed by the vehicle, and a mapping chart created by the creation unit representing the relationship between the position of the vehicle and the temperature of the control device among the types of the vehicle.

18. The cooling system according to claim 15, It is characterized in that The cooling execution unit starts cooling the control device from a predetermined time before the control device starts generating heat, using the prediction result of the temperature change of the control device acquired by the information acquisition unit.

19. The cooling system according to claim 15, It is characterized in that The cooling execution unit uses the prediction result of the temperature change of the control device acquired by the information acquisition unit to start cooling the control device using one or more cooling units selected from a plurality of cooling units according to the temperature change of the control device.

20. The cooling system according to claim 19, It is characterized in that The multiple cooling units include one or more air cooling units, one or more water cooling units, and multiple liquid nitrogen cooling units.

21. A cooling method, which is a cooling execution method executed by a server and a cooling execution device, It is characterized in that The cooling method comprises: an information acquisition step, wherein the server acquires, from the cooling execution device, position information of a vehicle having the cooling execution device and temperature information of a control device of the vehicle; a prediction step, wherein the server predicts a temperature change of the control device based on the position information of the vehicle having the cooling execution device and the temperature information of the control device of the vehicle acquired in the information acquisition step; an information acquisition step, wherein the cooling execution device acquires from the server a prediction result of the temperature change of the control device predicted by the prediction step of the server; and and a cooling execution step in which the cooling execution device starts cooling the control device based on the prediction of the temperature change of the control device acquired in the information acquisition step.

22. A cooling program, comprising a program executed by a server and a program executed by a cooling execution device, It is characterized in that The cooling process causes the server to perform the following steps: an information acquisition step of acquiring, from the cooling execution device, position information of a vehicle having the cooling execution device and temperature information of a control device of the vehicle; as well as a prediction step of predicting a temperature change of the control device based on the position information of the vehicle having the cooling execution device acquired in the information acquisition step and the temperature information of the control device of the vehicle, The cooling program causes the cooling execution device to execute the following steps: an information acquisition step of acquiring, from the server, a prediction result of the temperature change of the control device predicted by the prediction step of the server; as well as The cooling execution step starts cooling of the control device based on the prediction result of the temperature change of the control device acquired in the information acquisition step.

23. A cooling system comprising a server and a cooling execution device, It is characterized in that The server has: an information acquisition unit that acquires, from the cooling execution device, position information of a vehicle having the cooling execution device, temperature information of a control device of the vehicle, information related to traffic conditions, and weather information; as well as a prediction unit that predicts a temperature change of the control device based on the position information of the vehicle having the cooling execution device, the temperature information of the control device of the vehicle, the information related to the traffic conditions, and the weather information acquired by the information acquisition unit, The cooling execution device comprises: an information acquisition unit that acquires, from the server, a prediction result of a temperature change of the control device predicted by the prediction unit of the server; as well as A cooling execution unit starts cooling of the control device based on the prediction result of the temperature change of the control device acquired by the information acquisition unit.

24. The cooling system according to claim 23, It is characterized in that The server further includes a creation unit, which creates a mapping diagram representing the relationship between the position of the vehicle, the information related to the traffic condition, the weather information, and the temperature of the control device based on the position information of the vehicle having the cooling execution device, the temperature information of the control device of the vehicle, the information related to the traffic condition, and the weather information acquired by the information acquisition unit, The prediction unit predicts the temperature change of the control device based on the position information of the vehicle having the cooling execution device acquired by the information acquisition unit, the temperature information of the control device of the vehicle, the information related to the traffic condition, the weather information, and a mapping chart created by the creation unit that represents the relationship between the position of the vehicle, the information related to the traffic condition, the weather information and the temperature of the control device.

25. The cooling system according to claim 24, It is characterized in that The creation unit creates a mapping diagram representing the relationship between the position of the vehicle and the temperature of the control device according to each of the information related to the traffic condition and the weather information, based on the position information of the vehicle having the cooling execution device, the temperature information of the control device of the vehicle, the information related to the traffic condition, and the weather information.

26. The cooling system according to claim 24, It is characterized in that The information acquisition unit of the server further acquires information related to the type of the vehicle from the cooling execution device. The creation unit creates a mapping diagram representing the relationship between the position of the vehicle having the cooling execution device, the temperature information of the control device of the vehicle, the information related to the traffic condition, the weather information, the type of the vehicle and the temperature of the control device based on the position information of the vehicle having the cooling execution device, the temperature information of the control device of the vehicle, the information related to the traffic condition, the weather information, and the information related to the type of the vehicle.

27. The cooling system according to claim 23, It is characterized in that The cooling execution unit starts cooling the control device from a predetermined time before the control device starts generating heat, using the prediction result of the temperature change of the control device acquired by the information acquisition unit.

28. The cooling system according to claim 23, It is characterized in that The cooling execution unit uses the prediction result of the temperature change of the control device acquired by the information acquisition unit to start cooling the control device using one or more cooling units selected from a plurality of cooling units according to the temperature change of the control device.

29. The cooling system according to claim 28, It is characterized in that The multiple cooling units include one or more air cooling units, one or more water cooling units, and multiple liquid nitrogen cooling units.

30. A cooling method, which is a cooling execution method executed by a server and a cooling execution device, It is characterized in that The cooling method comprises: an information acquisition step, wherein the server acquires position information of a vehicle having the cooling execution device, temperature information of a control device of the vehicle, information related to traffic conditions, and weather information from the cooling execution device; and a prediction step, wherein the server predicts a temperature change of the control device based on the position information of the vehicle having the cooling execution device acquired in the information acquisition step, the temperature information of the control device of the vehicle, the information related to the traffic conditions, and the weather information, The cooling method further comprises: an information acquisition step, wherein the cooling execution device acquires from the server a prediction result of the temperature change of the control device predicted by the prediction step of the server; and and a cooling execution step in which the cooling execution device starts cooling the control device based on the prediction of the temperature change of the control device acquired in the information acquisition step.

31. A cooling program, comprising a program executed by a server and a program executed by a cooling execution device, It is characterized in that The cooling process causes the server to perform the following steps: an information acquisition step of acquiring, from the cooling execution device, position information of a vehicle having the cooling execution device and temperature information of a control device of the vehicle; as well as a prediction step of predicting a temperature change of the control device based on the position information of the vehicle having the cooling execution device acquired by the information acquisition step, the temperature information of the control device of the vehicle, the information related to the traffic conditions, and the weather information, The cooling program causes the cooling execution device to execute the following steps: an information acquisition step of acquiring, from the server, a prediction result of the temperature change of the control device predicted by the prediction step of the server; as well as The cooling execution step starts cooling of the control device based on the prediction result of the temperature change of the control device acquired in the information acquisition step.

32. A cooling system comprising a server and a cooling execution device, It is characterized in that The server has: an information acquisition unit that acquires, from the cooling execution device, position information of each integrated circuit in the control device, information related to processing of each integrated circuit, and information related to a driving condition of the vehicle; as well as a prediction unit that predicts temperature changes at various positions of the control device based on the position information of each integrated circuit in the control device, the information related to the processing of each integrated circuit, and the information related to the driving condition of the vehicle acquired by the information acquisition unit, The cooling execution device comprises: an information acquisition unit that acquires, from the server, a prediction result of a temperature change at each position of the control device predicted by the prediction unit of the server; as well as A cooling execution unit starts cooling of the control device based on the prediction result of the temperature change at each position of the control device acquired by the information acquisition unit.

33. The cooling system according to claim 32, It is characterized in that The prediction unit uses the position information of each integrated circuit in the control device acquired by the information acquisition unit, the information related to the processing of each integrated circuit, and the information related to the driving condition of the vehicle to predict the temperature change over time at each position of the control device.

34. The cooling system according to claim 32, It is characterized in that The cooling execution unit starts cooling of a position where heat is generated in the control device using the prediction result of the temperature change at each position of the control device acquired by the information acquisition unit.

35. The cooling system according to claim 34, It is characterized in that The cooling execution unit starts cooling a position where the control device generates heat from a predetermined time before the control device starts generating heat, using the prediction result of the temperature change at each position of the control device acquired by the information acquisition unit.

36. The cooling system according to claim 32, It is characterized in that The cooling execution unit uses the predicted results of the temperature changes at various positions of the control device obtained by the information acquisition unit, and uses one or more cooling units selected from a plurality of cooling units to start cooling the positions where heat is generated in the control device in response to the temperature changes at various positions of the control device.

37. The cooling system according to claim 36, It is characterized in that The multiple cooling units include one or more air cooling units, one or more water cooling units, and multiple liquid nitrogen cooling units.

38. A cooling method, performed by a server and a cooling execution device, It is characterized in that The cooling method comprises: an information acquisition step, wherein the server acquires position information of each integrated circuit in the control device, information related to processing of each integrated circuit, and information related to driving conditions of the vehicle from the cooling execution device; a prediction step, wherein the server predicts temperature changes at various positions of the control device based on the position information of each integrated circuit in the control device, the information related to processing of each integrated circuit, and the information related to the driving condition of the vehicle acquired in the information acquisition step; an information acquisition step, wherein the cooling execution device acquires from the server a prediction result of the temperature change at each position of the control device predicted by the prediction step of the server; and A cooling execution step in which the cooling execution device starts cooling the control device based on the prediction result of the temperature change at each position of the control device acquired in the information acquisition step.

39. A cooling program, comprising a program executed by a server and a program executed by a cooling execution device, It is characterized in that The cooling process causes the server to perform the following steps: an information acquisition step of acquiring, from the cooling execution device, position information of each integrated circuit in the control device, information related to processing of each integrated circuit, and information related to driving conditions of the vehicle; as well as a prediction step of predicting temperature changes at various positions of the control device based on the position information of each integrated circuit in the control device, the information related to the processing of each integrated circuit, and the information related to the driving condition of the vehicle acquired in the information acquisition step, The cooling program causes the cooling execution device to execute the following steps: an information acquisition step of acquiring, from the server, a prediction result of the temperature change at each position of the control device predicted by the prediction step of the server; as well as The cooling execution step starts cooling of the control device based on the prediction result of the temperature change at each position of the control device acquired in the information acquisition step.

40. A cooling execution device, comprising: a detection unit that detects a temperature of a control device that controls automatic driving of a vehicle and is mounted on the vehicle; and A cooling execution unit executes cooling of the control device using a given cooling means based on a period of time during which the temperature detected by the detection unit is continuously equal to or higher than a given temperature.

41. The cooling implementation device according to claim 40, It is characterized in that When the time for which the temperature remains above the given temperature exceeds a given threshold, the cooling execution unit executes cooling using the cooling units including one or more air cooling units, one or more water cooling units, and multiple liquid nitrogen cooling units.

42. The cooling implementation device according to claim 41, It is characterized in that The cooling execution unit executes rapid cooling when the temperature exceeds a given threshold value.

43. The cooling implementation device according to claim 40, It is characterized in that The cooling execution device further includes a prediction unit that predicts a temperature change of a control device that controls automatic driving of a vehicle and is mounted on the vehicle. The cooling execution unit executes cooling using the cooling unit based on the time during which the temperature remains at or above the predetermined temperature as the temperature change predicted by the prediction unit.

44. The cooling implementation device according to claim 41, It is characterized in that The cooling execution device further includes a prediction unit that predicts temperature changes of each of a plurality of parts of the control device. The cooling execution unit starts cooling the control device using a cooling unit selected from a plurality of cooling units that cool a plurality of locations of the control device, based on the prediction result of the prediction unit.

45. The cooling implementation device according to claim 44, It is characterized in that The control device has a plurality of processing chips respectively arranged at different positions of the control device. Each of the plurality of cooling units is disposed at a position corresponding to each of the plurality of processing chips.

46. ​​A cooling execution program for causing a computer to function as the cooling execution device according to any one of claims 40 to 45.

47. A cooling execution method, executed by a computer, The cooling execution method include: In a detection phase, a temperature of a control device is detected, wherein the control device controls the automatic driving of the vehicle and is mounted on the vehicle; as well as In the cooling execution phase, based on the time for which the temperature detected in the detection phase continues to be a given temperature or higher, cooling of the control device is executed using a given cooling unit.

48. A cooling execution device, It is characterized in that have: an estimating unit that estimates switching between automatic driving and manual driving based on a preset driving route; a creating unit that creates a cooling plan in which a given cooling unit is set based on the estimation result of the estimating unit; as well as A cooling execution unit executes cooling of a control device that controls automatic driving of a vehicle and is mounted on the vehicle, based on the cooling plan created by the creation unit.

49. The cooling implementation device according to claim 48, It is characterized in that The estimating unit estimates whether a switch from the automatic driving to the manual driving will occur based on a road condition of the driving route.

50. The cooling implementation device according to claim 48, It is characterized in that When the estimating unit estimates that the automatic driving is to be switched to the manual driving, the creating unit creates the cooling plan in which the cooling unit for the driving section of the manual driving is set. When the estimating unit estimates that the manual driving is to be switched to the automatic driving, the creating unit creates the cooling plan in which the cooling unit for the driving section of the automatic driving is set.

51. The cooling implementation device according to claim 48, It is characterized in that The cooling execution device also has: a detection unit that detects a temperature change of a control device that controls automatic driving of a vehicle and is mounted on the vehicle; and a prediction unit that predicts a temperature change of the control device, The creation unit compares the actual temperature change detected by the detection unit with the predicted temperature change predicted by the prediction unit, and creates the cooling plan in which the given cooling unit is set when the predicted temperature change is larger than the actual temperature change.

52. The cooling implementation device according to claim 48, It is characterized in that The cooling unit includes one or more air cooling units, one or more water cooling units, and multiple types of liquid nitrogen cooling units.

53. The cooling implementation device according to claim 52, It is characterized in that The cooling execution device further includes a prediction unit that predicts a temperature change of a control device that controls automatic driving of a vehicle and is mounted on the vehicle. The cooling execution unit executes cooling of the control device using a cooling unit selected from a plurality of cooling units that cool a plurality of locations of the control device, respectively, based on the cooling plan created by the creation unit.

54. The cooling implementation device according to claim 53, It is characterized in that The control device has a plurality of processing chips respectively arranged at different positions of the control device. Each of the plurality of cooling units is disposed at a position corresponding to each of the plurality of processing chips.

55. A program for causing a computer to function as the cooling execution device according to any one of claims 48 to 54.

56. A cooling execution method, executed by a computer, The cooling execution method include: In the estimation phase, the switch between automatic driving and manual driving is estimated based on the pre-set driving route; A creation phase, creating a cooling plan in which cooling conditions based on given cooling units are set based on the estimation result of the estimation phase; as well as In the cooling execution phase, cooling of a control device is executed based on the cooling plan created in the creation phase. The control device controls the automatic driving of the vehicle and is mounted on the vehicle.

57. A cooling device comprising: a prediction unit that predicts a temperature change of a control device that controls automatic driving of a vehicle and is mounted on the vehicle; and A cooling execution unit performs cooling so as to keep the temperature of the control device within a given temperature range based on the temperature change predicted by the prediction unit.

58. The cooling device according to claim 57, It is characterized in that The cooling execution unit determines the predetermined temperature range based on a threshold temperature preset for the control device.

59. The cooling device according to claim 57, It is characterized in that The prediction unit predicts the temperature change of the control device through artificial intelligence (AI).

60. The cooling device according to claim 59, It is characterized in that The cooling device also includes: a model storage unit storing a learning model generated by machine learning, which has the information acquired by the control device as an input and has a temperature change of the control device as an output, wherein the machine learning uses the information acquired by the control device and the temperature change of the control device when the control device acquires the information as learning data; as well as an information acquisition unit, which acquires the information acquired by the control device, The prediction unit predicts the temperature change of the control device by inputting the information acquired by the information acquisition unit into the learning model.

61. The cooling device according to claim 60, It is characterized in that The information acquisition unit acquires sensor information, which is acquired by the control device from a sensor mounted on the vehicle, from the sensor or the control device.

62. The cooling device according to claim 60, It is characterized in that The information acquisition unit acquires, from the control device, an analysis result of a captured image analyzed by the control device, the captured image being captured by a camera mounted on the vehicle.

63. The cooling device according to claim 60, It is characterized in that The information acquisition unit acquires external information received by the control device from an external device from the external device or the control device.

64. The cooling device according to claim 63, It is characterized in that The information acquisition unit acquires the traffic information of the road on which the vehicle is located, which is received by the control device from the external device, from the external device or the control device.

65. The cooling device according to claim 57, It is characterized in that The cooling execution unit starts cooling the control device using one or more cooling units selected from a plurality of cooling units, depending on the level of the temperature of the control device predicted by the prediction unit.

66. The cooling device according to claim 65, It is characterized in that The multiple cooling units include one or more air cooling units, one or more water cooling units, and multiple liquid nitrogen cooling units.

67. The cooling device according to claim 66, It is characterized in that The prediction unit predicts temperature changes of each of the plurality of parts of the control device. The cooling execution unit starts cooling the control device using a cooling unit selected from a plurality of cooling units that cool a plurality of locations of the control device, based on the prediction result of the prediction unit.

68. The cooling device according to claim 67, It is characterized in that The control device has a plurality of processing chips respectively arranged at different positions of the control device. Each of the plurality of cooling units is disposed at a position corresponding to each of the plurality of processing chips.

69. A cooling method, performed by a cooling device, It is characterized in that The cooling method comprises: a prediction step of predicting a temperature change of a control device that controls automatic driving of a vehicle and is mounted on the vehicle; and The cooling execution step performs cooling so as to keep the temperature of the control device within a predetermined temperature range based on the temperature change predicted by the prediction step.

70. A cooling program for causing a computer to perform the following process: A prediction process for predicting a temperature change of a control device that controls automatic driving of a vehicle and is mounted on the vehicle; and The cooling execution process performs cooling so as to keep the temperature of the control device within a given temperature range based on the temperature change predicted in the prediction process.

71. A cooling execution device, It is characterized in that have: a detection unit that detects a temperature change of a control device that controls automatic driving of a vehicle and is mounted on the vehicle; a selection unit that selects a given operating condition for lowering the temperature of the control device when the temperature change exceeds a given threshold value; as well as An output unit outputs the given operating condition based on the selection result of the selection unit.

72. The cooling implementation device according to claim 71, It is characterized in that When the temperature change exceeds a predetermined threshold value, the selection unit selects, as the predetermined operating condition, an operating condition that reduces the amount of calculation involved in predetermined information processing of the control device.

73. The cooling implementation device according to claim 71, It is characterized in that The cooling execution device also has an information acquisition unit, which acquires the result of information processing related to the automatic driving from an external information processing device, and the external information processing device performs the information processing related to the automatic driving as the given information processing of the control device.

74. The cooling implementation device according to claim 71, It is characterized in that When the temperature change exceeds a given threshold value, the selection unit selects, as the given operating condition, an operating condition for suppressing the driving speed of the automatic driving to be equal to or lower than a given speed.

75. The cooling implementation device according to claim 71, It is characterized in that The selection unit selects, as the predetermined operating condition, an operating condition for changing the automatic driving to the manual driving when the temperature change exceeds a predetermined threshold value.

76. The cooling implementation device according to claim 71, It is characterized in that The selection unit selects, as the predetermined operating condition, an operating condition of stopping the vehicle at a predetermined position on a road included in the autonomous driving driving route when the temperature change exceeds a predetermined threshold value.

77. The cooling implementation device according to claim 71, It is characterized in that When the temperature change exceeds a predetermined threshold value, the selection unit selects, as the predetermined operating condition, an operating condition that suppresses acquisition of predetermined information used in information processing for the autonomous driving.

78. A program for causing a computer to function as the cooling execution device according to any one of claims 71 to 77.

79. A cooling execution method, executed by a computer, the cooling execution method include: In a detection phase, a temperature change of a control device is detected, wherein the control device controls the automatic driving of the vehicle and is mounted on the vehicle; In a selection stage, when the temperature change exceeds a given threshold, a given operating condition is selected to reduce the temperature of the control device; as well as The output stage outputs the given operating condition based on the selection result of the selection stage.

80. A cooling device, It is characterized in that have: a prediction unit that predicts a temperature change of a control device that controls automatic driving of a vehicle and is mounted on the vehicle; a communication unit that communicates with other vehicles existing within a given range based on the temperature change predicted by the prediction unit; as well as An instruction unit instructs the control device to execute a calculation related to the control of the automatic driving of another vehicle communicated with by the communication unit.

81. The cooling device according to claim 80, It is characterized in that The communication unit determines that another vehicle existing in a range in which communication with another vehicle can be maintained for a predetermined time is a vehicle existing in the predetermined range, and performs communication.

82. The cooling device according to claim 80, It is characterized in that The cooling device further includes a cooling execution unit configured to start cooling of the control device based on the temperature change predicted by the prediction unit.

83. The cooling device according to claim 82, It is characterized in that The cooling execution unit starts cooling the control device in response to the prediction by the prediction unit that the temperature of the control device will become higher than a preset threshold value.

84. The cooling device according to claim 80, It is characterized in that The prediction unit predicts the temperature change of the control device through artificial intelligence (AI).

85. The cooling device according to claim 84, It is characterized in that The cooling device also includes: a model storage unit storing a learning model generated by machine learning, which has the information acquired by the control device as an input and has a temperature change of the control device as an output, wherein the machine learning uses the information acquired by the control device and the temperature change of the control device when the control device acquires the information as learning data; as well as an information acquisition unit, which acquires the information acquired by the control device, The prediction unit predicts the temperature change of the control device by inputting the information acquired by the information acquisition unit into the learning model.

86. The cooling device according to claim 85, It is characterized in that The information acquisition unit acquires sensor information, which is acquired by the control device from a sensor mounted on the vehicle, from the sensor or the control device.

87. The cooling device according to claim 85, It is characterized in that The information acquisition unit acquires, from the control device, an analysis result of a captured image analyzed by the control device, the captured image being captured by a camera mounted on the vehicle.

88. The cooling device according to claim 85, It is characterized in that The information acquisition unit acquires external information received by the control device from an external device from the external device or the control device.

89. The cooling device according to claim 88, It is characterized in that The information acquisition unit acquires the traffic information of the road on which the vehicle is located, which is received by the control device from the external device, from the external device or the control device.

90. The cooling device according to claim 82, It is characterized in that The cooling execution unit starts cooling the control device using one or more cooling units selected from a plurality of cooling units, depending on the level of the temperature of the control device predicted by the prediction unit.

91. The cooling device according to claim 90, It is characterized in that The multiple cooling units include one or more air cooling units, one or more water cooling units, and multiple liquid nitrogen cooling units.

92. The cooling device according to claim 91, It is characterized in that The prediction unit predicts temperature changes of each of the plurality of parts of the control device. The cooling execution unit starts cooling the control device using a cooling unit selected from a plurality of cooling units that cool a plurality of locations of the control device, based on the prediction result of the prediction unit.

93. The cooling device according to claim 92, It is characterized in that The control device has a plurality of processing chips respectively arranged at different positions of the control device. Each of the plurality of cooling units is disposed at a position corresponding to each of the plurality of processing chips.

94. A cooling method, performed by a cooling device, It is characterized in that The cooling method comprises: A prediction step of predicting a temperature change of a control device that controls automatic driving of a vehicle and is mounted on the vehicle; a communication step of communicating with other vehicles existing within a given range based on the temperature change predicted by the prediction step; and An instruction step of instructing the control device to execute a calculation related to the control of the automatic driving of the other vehicle communicated with in the communication step.

95. A cooling program for causing a computer to perform the following process: A prediction process predicts a temperature change of a control device, the control device controls automatic driving of a vehicle and is mounted on the vehicle; a communication process for communicating with other vehicles existing within a given range based on the temperature change predicted by the prediction process; as well as An instruction process for instructing the control device to execute operations related to the control of the automatic driving of another vehicle communicated through the communication process.

96. A cooling execution device, comprising: a prediction unit that predicts changes in computing power of a control device that controls automatic driving of a vehicle and is mounted on the vehicle; and A cooling execution unit starts cooling of the control device based on the change predicted by the prediction unit.

97. The cooling implementation device according to claim 96, in, The prediction unit predicts changes in the computing power of the control device through artificial intelligence (AI).

98. The cooling implementation device according to claim 97, in, The prediction unit predicts a moment when the computing power of the control device becomes maximum, The cooling execution unit starts cooling of the control device based on the instant predicted by the prediction unit.

99. The cooling execution device according to claim 98, in, The cooling execution unit starts cooling the control device at the moment when the computing power of the control device becomes maximum.

100. The cooling execution device according to claim 99, in, The prediction unit predicts the time until the temperature of the control device becomes higher than a preset temperature based on the prediction result of the instant when the computing power of the control device becomes maximum, The cooling execution unit starts cooling of the control device based on the time estimated by the estimation unit.

101. The cooling execution device according to claim 96, in, The cooling execution device also includes: a model storage unit storing a learning model generated by machine learning, which takes the information acquired by the control device as input and takes the computing power of the control device as output, wherein the machine learning takes the information acquired by the control device and the computing power of the control device when the control device acquires the information as learning data; as well as an information acquisition unit, which acquires the information acquired by the control device, The prediction unit predicts a change in the computing power of the control device by inputting the information acquired by the information acquisition unit into the learning model.

102. The cooling execution device according to claim 101, in, The information acquisition unit acquires at least any one of sensor information acquired by the control device from a sensor mounted on the vehicle, analysis results of an image captured by a camera mounted on the vehicle after the control device has analyzed the image, external information received by the control device from an external device, and traffic information of the road on which the vehicle is located received by the control device from the external device.

103. The cooling execution device according to claim 96, in, The cooling execution device also includes: a model storage unit storing a learning model generated by machine learning, which has information acquired by the control device as input and has a time until the control device is cooled when cooling of the control device is started as output, wherein the machine learning has the information acquired by the control device and the time until the control device is cooled when cooling of the control device is started when the control device acquires the information as learning data; as well as an information acquisition unit, which acquires the information acquired by the control device, The cooling execution unit determines the timing to start cooling the control device based on the change predicted by the prediction unit and the time from when cooling of the control device is started to when the control device is cooled, which is obtained by inputting the information acquired by the information acquisition unit into the learning model.

104. The cooling implementation device according to claim 96, in, The cooling execution unit adjusts the degree of cooling of the control device according to a risk rate, wherein the risk rate represents the probability of occurrence of a risk that the control device becomes hot and thus the control device no longer operates normally, thereby affecting normal autonomous driving of the vehicle.

105. The cooling execution device according to claim 104, in, When the risk rate is lower than a first threshold value, the cooling execution unit cools the control device with a first cooling intensity; when the risk rate is higher than the first threshold value and lower than a second threshold value higher than the first threshold value, the cooling execution unit cools the control device with a second cooling intensity stronger than the first cooling intensity; when the risk rate is higher than the second threshold value, the cooling execution unit cools the control device with a third cooling intensity stronger than the second cooling intensity.

106. A program for causing a computer to function as the cooling execution device according to any one of claims 96 to 104.

107. A cooling execution method, executed by a computer, the cooling execution method include: In a prediction phase, a change in the computing power of a control device is predicted, the control device controls the automatic driving of the vehicle and is mounted on the vehicle; as well as In the cooling execution phase, cooling of the control device is started based on the change predicted in the prediction phase.

Images