Travelling route management system

Abstract

To secure transparency and easiness in verification of automatic drive control when storing log data related to the automatic drive control for a vehicle.SOLUTION: A traveling route management system generates a travelling route in an automatic drive control for a vehicle. The travelling route management system includes: a storage device storing a travelling history managing a process where the vehicle traveled by the automatic drive control in the past; and one or more processors. The vehicle includes a target storage device storing log data related to the aromatic drive control. One or more processors are constituted so as to execute: processing for acquiring residual capacity of the target storage device; processing for generating one or more candidate routes; and processing for selecting the travelling route in one or more candidate routes. The processing for selecting the travelling route comprises selecting the travelling route among the one or more candidate routes while using a large number of common points with the travelling history as an index when the residual capacity of the target storage device is smaller than a prescribed capacity.SELECTED DRAWING: Figure 3

Description

[Technical field]

[0001] The present disclosure relates to a route management system that generates routes for autonomous driving control. [Background technology]

[0002] Patent Document 1 discloses a driving information management device that can record routes traveled in the past under automatic driving as historical routes. The driving information management device is mounted on a vehicle equipped with an automatic driving device that can switch between manual driving and automatic driving. According to the driving information management device, when the automatic driving device is switched to automatic driving, the generation of automatic driving history data that indicates the history of driving routes where automatic driving was performed is automatically started. The automatic driving history data is stored in a database, and when the vehicle is traveling, a route that can be automatically driven can be called up from the past history.

[0003] In addition to the above-mentioned Patent Document 1, the following Patent Documents 2 and 3 can be mentioned as examples of documents showing the state of the art in the technical field related to the present disclosure. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2020-060901 A [Patent Document 2] JP 2017-134725 A [Patent Document 3] JP 2020-008441 A Summary of the Invention [Problem to be solved by the invention]

[0005] Technologies for autonomous driving of vehicles are known. When a vehicle is driven autonomously, it is desirable to store data related to the autonomous driving control so that how the autonomous driving of the vehicle was performed can be verified later. However, depending on the degree of data capacity shortage, it may be difficult to store all the data. In such cases, the issue is how to ensure the transparency and ease of verification of the autonomous driving control.

[0006] One objective of the present disclosure is to provide a technology that enables ensuring the transparency and ease of verification of autonomous driving control when storing log data related to autonomous driving control of a vehicle. [Means for solving the problem]

[0007] The present disclosure provides a driving route management system that generates driving routes for automatic driving control of a vehicle. The driving route management system includes a storage device that stores a driving history that manages the routes that a vehicle has traveled in the past under automatic driving control, and one or more processors. The vehicle is equipped with a target storage device that stores log data related to autonomous driving control. The one or more processors are configured to execute a process of acquiring remaining capacity of a target storage device, a process of generating one or more candidate routes, and a process of selecting a driving route from the one or more candidate routes. The process of selecting a driving route includes, when the remaining capacity of the target storage device is less than a predetermined amount, selecting a driving route from one or more candidate routes using the number of commonalities with the driving history as an index. Effect of the Invention

[0008] According to the technology disclosed herein, a driving route management system includes a storage device that stores a driving history that manages the routes that a vehicle has traveled in the past under automatic driving control. The driving route management system first generates candidate routes that are candidates for driving routes. When the remaining capacity of the storage device that stores log data related to automatic driving control is equal to or less than a predetermined amount, a driving route is selected from the candidate routes using the number of commonalities with the driving history as an indicator. When traveling along the driving route generated in this manner, a large portion of the data related to automatic driving control can be supplemented by previously stored log data without being subject to storage.

[0009] In this way, according to the present disclosure, it is possible to ensure the transparency and ease of verification of autonomous driving control. [Brief description of the drawings]

[0010] [Figure 1] FIG. 2 is a diagram showing a configuration example related to automatic driving control of a vehicle according to the present embodiment. [Diagram 2] FIG. 1 is a diagram illustrating an example of a configuration of an automatic driving system according to an embodiment of the present invention. [Diagram 3] 10 is a flowchart showing an example of a process for generating a travel route. [Figure 4] 13 is a flowchart illustrating an example of a process related to saving control data. [Diagram 5] 13 is a flowchart showing an example of a process for generating a travel route in a first modified example. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0011] 1.Autonomous driving of vehicles 1 is a block diagram showing an example of a configuration related to the automatic driving control of a vehicle 1 according to the present embodiment. Automatic driving refers to automatically performing at least one of steering, acceleration, and deceleration of the vehicle 1 without the driver's operation. Automatic driving control is a concept that includes not only fully automatic driving control, but also risk avoidance control, lane keep assist control, and the like. The operator may be a driver who rides in the vehicle 1, or a remote operator who remotely operates the vehicle 1.

[0012] The vehicle 1 includes a sensor group 10, a recognition unit 20, a planner 30, a control amount calculation unit 40, a travel route management unit 50, and a travel device 60.

[0013] The sensor group 10 includes a recognition sensor 11, a status sensor 12, and a position sensor 13. The recognition sensor 11 is used to recognize the surrounding situation of the vehicle 1. Examples of the recognition sensor 11 include a camera, a LIDAR (Laser Imaging Detection and Ranging), a radar, and the like. The status sensor 12 detects the status of the vehicle 1. Examples of the status sensor 12 include a speed sensor, an acceleration sensor, a yaw rate sensor, a steering angle sensor, and the like. The position sensor 13 detects the position of the vehicle 1. Examples of the position sensor 13 include a GNSS (Global Navigation Satellite System) sensor.

[0014] The sensor detection information SEN is information obtained by the sensor group 10. The sensor detection information SEN includes recognition sensor information, vehicle status information, and position information. The recognition sensor information is information obtained by the recognition sensor 11 recognizing the surroundings of the vehicle 1. For example, the recognition sensor information includes images captured by a camera and point cloud information obtained by LIDAR. The vehicle status information is information indicating the status of the vehicle 1, obtained by the status sensor 12. The position information is information indicating the position of the vehicle 1, and is obtained from the position sensor 13.

[0015] The recognition unit 20, the planning unit 30, and the control amount calculation unit 40 constitute an "automatic driving control unit" that controls the automatic driving of the vehicle 1. The recognition unit 20 receives sensor detection information SEN including at least recognition sensor information from the sensor group 10. The recognition unit 20 recognizes the situation around the vehicle 1 based on the received sensor detection information SEN. For example, the recognition unit 20 recognizes objects around the vehicle 1. Examples of objects include pedestrians, other vehicles (preceding vehicles, parked vehicles, etc.), white lines, road structures (e.g., guardrails, curbs), fallen objects, traffic lights, intersections, signs, etc. The recognition result information RES indicates the recognition result by the recognition unit 20. For example, the recognition result information RES includes object information indicating the relative position and relative speed of the object with respect to the vehicle 1, and information regarding the curvature and road width of the road on which the vehicle 1 is traveling.

[0016] The driving route management unit 50 generates and manages a driving route that indicates the route along which the vehicle 1 will travel to the destination. The driving route may include a designation of the lanes along which the vehicle 1 will travel. The generation of the driving route by the driving route management unit 50 will be described later.

[0017] The planner 30 receives the recognition result information RES from the recognition unit 20. The planner 30 also receives the driving route from the driving route management unit 50. The planner 30 may also receive vehicle state information, position information, and map information generated in advance. The map information may be high-precision three-dimensional map information. The planner 30 generates a driving plan for driving along the driving route based on the received information. Examples of the driving plan include maintaining the current driving lane, changing lanes, overtaking, turning right or left, steering, accelerating, decelerating, and stopping. Furthermore, the planner 30 generates a target trajectory TRJ required for the vehicle 1 to drive according to the driving plan. The target trajectory TRJ includes a target position and a target speed. The driving plan and the target trajectory TRJ may include those generated to avoid risks.

[0018] The control amount calculation unit 40 receives the target trajectory TRJ from the planning unit 30. The control amount calculation unit 40 calculates a control amount CON required for the vehicle 1 to follow the target trajectory TRJ. It can also be said that the control amount CON is a control amount required to reduce the deviation between the vehicle 1 and the target trajectory TRJ. The control amount CON includes at least one of a steering control amount, a drive control amount, and a braking control amount. Examples of the steering control amount include a target steering angle, a target torque, a target motor angle, a target motor drive current, etc. Examples of the drive control amount include a target speed, a target acceleration, etc. Examples of the braking control amount include a target speed, a target deceleration, etc.

[0019] The traveling device 60 includes a steering device 61, a driving device 62, and a braking device 63. The steering device 61 steers the wheels. For example, the steering device 61 includes an electric power steering (EPS) device. The driving device 62 is a power source that generates a driving force. Examples of the driving device 62 include an engine, an electric motor, and an in-wheel motor. The braking device 63 generates a braking force. The traveling device 60 receives a control amount CON from the control amount calculation unit 40. The traveling device 60 operates the steering device 61, the driving device 62, and the braking device 63 according to the steering control amount, the driving control amount, and the braking control amount, respectively. As a result, the vehicle 1 travels so as to follow the target trajectory TRJ.

[0020] 2 is a conceptual diagram showing a configuration example of an autonomous driving system 100 according to this embodiment. The autonomous driving system 100 is mounted on a vehicle 1 and controls autonomous driving of the vehicle 1. The autonomous driving system 100 has at least the functions of the autonomous driving control unit described above. The autonomous driving system 100 may further include a sensor group 10 and a driving device 60.

[0021] The autonomous driving system 100 includes one or more processors 110 (hereinafter simply referred to as processors 110) and one or more memory devices 120 (hereinafter simply referred to as memory devices 120).

[0022] The processor 110 executes various processes. Examples of the processor 110 include a central processing unit (CPU), a graphics processing unit (GPU), an application specific integrated circuit (ASIC), and a field-programmable gate array (FPGA).

[0023] The storage device 120 stores a program 130 and various information. Examples of the storage device 120 include a hard disk drive (HDD), a solid state drive (SSD), a volatile memory, and a non-volatile memory.

[0024] The program 130 is a computer program for controlling the vehicle 1, and is executed by the processor 110. The program 130 may be recorded in a computer-readable recording medium. Various processes by the autonomous driving system 100 are realized by cooperation between the processor 110 that executes the program 130 and the storage device 120. In other words, the recognition unit 20, the plan unit 30, the control amount calculation unit 40, and the driving path management unit 50 are realized by the processor 110 executing the program 130. Each functional unit may be realized by a single processor 110, or may be realized by separate processors 110.

[0025] The various information stored in the storage device 120 includes map information and a driving history DAT. The driving history DAT is data that manages the routes that the vehicle 1 has traveled in the past under automatic driving control. Examples of the driving history DAT include data on routes that the vehicle 1 has traveled in the past, data on lanes that the vehicle 1 has traveled in the past, and information on the road shapes within the routes that the vehicle 1 has traveled in the past. Furthermore, the driving history DAT may include information on the time periods in which the vehicle 1 has traveled in the past.

[0026] The data on the route traveled in the past by the vehicle 1 is, for example, a map showing the route traveled in the past by the vehicle 1. The information on the road shape in the route traveled in the past by the vehicle 1 is, for example, information showing the curvature, gradient, road width, etc. of at least a part of the route traveled in the past by the vehicle 1. Alternatively, it may be information showing whether a specific road shape was included in the route traveled in the past by the vehicle 1. Examples of the specific road shape include a curved road with a predetermined curvature, a ramp with a predetermined gradient, a branch, a junction, road markings, a road under construction, a road separated by something other than lanes such as a ball or a pylon, etc.

[0027] Referring again to FIG. 1, the recognition unit 20 includes at least one of a rule-based model and a machine learning model. The rule-based model performs recognition processing based on a group of predetermined rules. Examples of the machine learning model include a Neural Network (NN), a Support Vector Machine (SVM), a regression model, a decision tree model, and the like. The NN may be a Convolutional Neural Network (CNN), a Recurrent Neural Network (RNN), or a combination thereof. The type of each layer, the number of layers, and the number of nodes in the NN are arbitrary. The machine learning model is generated in advance through machine learning. The recognition unit 20 performs recognition processing by inputting the sensor detection information SEN to the model. The recognition result information RES is output from the model, or is generated based on the output from the model.

[0028] Similarly, the planner 30 includes at least one of a rule-based model and a machine learning model. The planner 30 performs planning processing by inputting the recognition result information RES to the model. The target trajectory TRJ is output from the model or is generated based on the output from the model.

[0029] Similarly, the control amount calculation unit 40 includes at least one of a rule-based model and a machine learning model. The control amount calculation unit 40 performs a control amount calculation process by inputting a target trajectory TRJ to the model. The control amount CON is output from the model or is generated based on the output from the model. The storage device 120 stores data of the recognition unit 20, the planning unit 30, and the models included in the control amount calculation unit 40, and is used for automatic driving control.

[0030] Two or more of the recognition unit 20, the planner 30, and the control amount calculation unit 40 may be integrally configured. All of the recognition unit 20, the planner 30, and the control amount calculation unit 40 may be integrally configured (end-to-end configuration). For example, the recognition unit 20 and the planner 30 may be integrally configured by a NN that outputs a target trajectory TRJ from the sensor detection information SEN. Even in the case of an integral configuration, intermediate products such as recognition result information RES and target trajectory TRJ may be output. For example, when the recognition unit 20 and the planner 30 are integrally configured by a NN, the recognition result information RES may be the output of an intermediate layer of the NN.

[0031] In this embodiment, a machine learning model is used for at least a part of the autonomous driving control unit. That is, at least one of the recognition unit 20, the planner 30, and the control amount calculation unit 40 includes a machine learning model. The autonomous driving control unit performs at least a part of the autonomous driving control of the vehicle 1 by using the machine learning model.

[0032] During the autonomous driving control, the processor 110 acquires a log of data related to the autonomous driving control. Hereinafter, this data is referred to as control data. The processor 110 stores a log of some or all of the control data acquired during the autonomous driving control in the storage device 120 as log data LOG. The processor 110 may store the control data in the storage device 120 for only a certain period of time. The storage of the control data is performed to ensure the explainability of the autonomous driving control.

[0033] As described above, the autonomous driving system 100 according to the present embodiment is configured. In the autonomous driving system 100, the processor 110 that realizes the driving route management unit 50 and the storage device 120 that stores the driving history DAT realize the driving route management system according to the present embodiment. Note that the storage device 120 may include a storage device for storing the driving history DAT and a storage device for storing the log data LOG separately.

[0034] 2. Process for generating driving routes Fig. 3 is a flowchart showing an example of processing related to generation of a driving route, which is executed in the driving route management unit 50. The processing according to the flowchart shown in Fig. 3 is executed at the timing when the autonomous driving system 100 determines the driving route of the vehicle 1. The timing for determining the driving route of the vehicle 1 is, for example, the timing when the vehicle 1 starts driving by autonomous driving. Alternatively, the driving route may be changeable at any timing after the vehicle 1 starts driving, and the processing may be executed at the timing when the driving route is changed.

[0035] In step S110, the processor 110 generates a candidate route. The candidate route is a route that is a candidate for the travel route. The processor 110 acquires, for example, the destination input by the operator, the position information of the vehicle 1 included in the sensor detection information SEN, and the map information stored in the storage device 120, and generates a route connecting the current position of the vehicle 1 and the destination as the candidate route. There may be multiple candidate routes. Furthermore, the candidate route may include, in addition to the route on which the vehicle 1 travels, information specifying the lanes on which the vehicle 1 travels within the route.

[0036] In step S120, the processor 110 acquires the remaining capacity of the storage device 120. If the storage device 120 includes a plurality of storage devices, the remaining capacity of the storage device for storing the log data LOG is acquired here.

[0037] In step S130, processor 110 determines whether the remaining capacity acquired in step S120 is less than a predetermined amount. If the remaining capacity is less than the predetermined amount (step S130; Yes), the process proceeds to step S140. On the other hand, if the remaining capacity is equal to or greater than the predetermined amount (step S130; No), the process proceeds to step S160.

[0038] In step S140, the processor 110 selects a travel route from the candidate routes generated in step S110. At this time, the processor 110 uses the number of commonalities with the travel history DAT as an index and selects the candidate route having many commonalities as the travel route.

[0039] The number of commonalities between the candidate route and the driving history DAT will be described. The number of commonalities is determined, for example, by the number of common parts of the routes. The processor 110 acquires routes that the vehicle 1 has traveled in the past from the driving history DAT and compares them with the candidate route. Then, it is determined that the more routes that the vehicle 1 has traveled in the past, the more commonalities there are between the candidate route and the vehicle 1. The more routes that the vehicle 1 has traveled in the past include, the more commonalities there are between the candidate route and the vehicle 1. For example, the candidate route includes a large proportion of routes that overlap with routes that the vehicle 1 has traveled in the past, or the length of the portion excluding the routes that the vehicle 1 has traveled in the past is short.

[0040] When the candidate route includes information specifying the lane on which the vehicle 1 travels within the route, the number of commonalities with the driving history DAT may be determined based on the number of commonalities of the lanes on which the vehicle 1 travels. For example, the processor 110 acquires information on the route on which the vehicle 1 has traveled in the past and which lanes on the route the vehicle 1 traveled on from the driving history DAT. Then, the processor 110 determines that the more lanes the candidate route includes on which the vehicle 1 has traveled in the past, the more commonalities the candidate route has with the driving history DAT. At this time, the processor 110 may also determine that the lanes are common for candidate routes that travel on the same lanes of different routes. For example, the driving history DAT includes only data on the vehicle 1 traveling on the left lane in the past, and does not include data on the vehicle 1 traveling on the right lane. In this case, the more parts of the candidate route that specify the left lane as the lane on which the vehicle 1 travels, the more commonalities the candidate route may be determined to have.

[0041] If the candidate route does not include a route that the vehicle 1 has traveled in the past, the number of commonalities between the candidate route and the travel history DAT may be determined based on the number of commonalities of the road shapes. In this case, the processor 110 acquires information on the road shapes included in each of the candidate routes from, for example, map information stored in the storage device 120. Then, the more of the candidate routes that include road shapes common to road shapes that the vehicle 1 has experienced in the past, the more commonalities between the candidate route and the travel history DAT are determined. Here, the fact that the vehicle 1 has experienced a specific road shape means that the specific road shape was present on a route that the vehicle 1 has traveled in the past. For example, it is assumed that the vehicle 1 has experienced a merging in the past, but has not experienced a slope with a predetermined gradient. Then, it is assumed that, as a result of acquiring information on the road shapes included in each of the candidate routes, some of the candidate routes include a merging and others include a slope with a predetermined gradient. In this case, the processor 110 may determine that the candidate route that includes a merging has more commonalities.

[0042] As an example of a candidate route, consider a case where the following five routes are included. The first is a candidate route in which all routes from the departure point to the destination overlap with routes included in the driving history DAT and also share the same lanes. The second is a candidate route in which all routes are common to the driving history DAT, but the lanes are not common. The third is a candidate route in which the proportion of common routes is high, although some routes are not common to the driving history DAT. The fourth example is a candidate route in which the proportion of common routes is high, but which does not include routes common to the driving history DAT. The fifth example is a candidate route in which the proportion of common routes is low, but which does not include routes common to the driving history DAT. The fifth is a candidate route in which the proportion of common routes is low, but which does not include routes common to the driving history DAT. In this case, it is determined that the first candidate route has more common points, and the fifth candidate route has fewer common points.

[0043] If the remaining capacity is equal to or greater than the predetermined amount, in step S160, the processor 110 selects a travel route from the candidate routes generated in step S130. The travel route is selected from one or more candidate routes in the same manner as in step S140. However, when the travel route is selected in step S160, the number of common points with the travel history DAT is not used as an index for selecting the travel route.

[0044] The method of selecting the travel route in step S160 is arbitrary. For example, the processor 110 may select the shortest route among the candidate routes as the travel route, may select the route with the smallest gradient as the travel route, or may select the travel route from the viewpoint of fuel efficiency.

[0045] When a travel route is selected in step S140 or step S160, the series of processes ends. In this manner, a travel route is generated by the series of processes. The generated travel route is acquired by the planning unit 30, and the planning unit 30 generates a travel plan for the vehicle 1 to travel along the travel route. If a travel route has already been generated when the series of processes ends, the travel route is changed, and the planning unit 30 generates a new travel plan for traveling along the new travel route.

[0046] 3. Processing related to storage of control data As described above, while the automatic driving control is being performed, the processor 110 acquires control data and stores it as log data LOG in the storage device 120. Fig. 4 is a flowchart showing a process related to the storage of the control data. The process according to the flowchart shown in Fig. 4 is repeatedly executed at a predetermined processing cycle, for example, while the automatic driving control is being performed.

[0047] In step S210, the processor 110 acquires data related to the automatic driving control. Here, assuming that the data will be used for later verification, etc., basically, all data used in the automatic driving control is acquired. For example, the sensor detection information SEN input to the recognition unit 20 and the recognition result information RES output from the recognition unit 20 are acquired. Alternatively, for example, when the recognition unit 20, the planner 30, and the control amount calculation unit 40 are integrally configured, the sensor detection information SEN input to the automatic driving control unit, the recognition result information RES and the target trajectory TRJ as intermediate products, and the control amount CON output from the automatic driving control unit are acquired. The acquired data may further include the reason for the judgment in the recognition process by the recognition unit 20, the reason for the judgment in the planner 30, the presence or absence of operator intervention in the automatic driving control, etc.

[0048] In step S220, the processor 110 acquires the remaining capacity of the storage device 120. The size of the remaining capacity of the storage device 120 acquired here means the size of the capacity in which new control data can be stored. Note that, in the case where the storage device 120 includes multiple storage devices, the remaining capacity of the storage device for storing the log data LOG is acquired here.

[0049] In step S230, processor 110 determines whether the remaining capacity acquired in step S210 is less than a predetermined amount. The predetermined amount here may be greater than or equal to the predetermined amount in step S130 of Fig. 3. If the remaining capacity is less than the predetermined amount (step S230; Yes), the process proceeds to step S240.

[0050] In step S240, the processor 110 stores the difference data in the storage device 120. The difference data is data that is the difference between the data acquired in step S210 and the log data LOG. For example, when the vehicle 1 is traveling on a route that has been traveled in the past, it is expected that the surrounding scenery captured by the camera in the sensor detection information SEN input to the recognition unit 20 and the detection result of the position of the white line in the recognition result information RES output by the recognition unit 20 will be approximately equivalent to the data included in the previously stored log data LOG. Such data excluding the data equivalent to the log data LOG becomes the difference data. In other words, the storage device 120 newly stores the data acquired in step S210 excluding the part that can be complemented by the previously stored log data LOG.

[0051] On the other hand, if the remaining capacity is equal to or greater than the predetermined amount (step S230; No), the process proceeds to step S250. In step S250, processor 110 stores in storage device 120 all of the data acquired in step S210.

[0052] The data stored in storage device 120 in step S240 or step S250 is control data. When the control data is stored in storage device 120, the series of processes ends.

[0053] 4.Effects The effects of these processes will be described below. As described above, the control data is stored in preparation for the case where it becomes necessary to verify later how the automatic driving control was performed. Therefore, it is desirable to store as much control data as possible acquired while the automatic driving control is being performed, at least until the verification is completed and it is determined that storage is not necessary.

[0054] However, since the capacity of the storage device 120 is limited, the capacity of the control data that can be stored in the storage device 120 is also limited. If the remaining capacity of the storage device 120 decreases and falls short of the required capacity, new control data cannot be stored in the storage device 120. Therefore, when the remaining capacity of the storage device 120 decreases and it is suggested that the capacity may be tight, it is necessary to conserve the remaining capacity.

[0055] Here, according to the automatic driving system 100 according to the present embodiment, when the remaining capacity of the storage device 120 falls below a predetermined amount, a driving route is generated that has many commonalities with the routes traveled by the vehicle 1 under automatic driving control in the past. When traveling along the driving route generated in this way, a large portion of the control data can be supplemented by the log data LOG stored in the past. This makes it possible to ensure the transparency and ease of verification of the automatic driving control even in cases where it is difficult to store the control data.

[0056] Furthermore, according to this embodiment, the control data stored in the storage device 120 is differential data.

[0057] The differential data is data related to autonomous driving control, excluding parts that overlap with previously stored log data LOG. Since the overlapping parts can be supplemented by past data, if at least the differential data is stored, the explainability of the situation can be guaranteed by combining the previously acquired data with the currently acquired differential data. In other words, the differential data can be said to be the minimum data that should be stored assuming that it will be used for future verification. When the remaining capacity is low, the amount of newly stored control data can be reduced by limiting the data to be stored to the minimum necessary data and omitting storage of parts that can be explained by combining with past data.

[0058] In addition, the selection of a candidate route having many commonalities with the driving history DAT as the driving route is effective in reducing the capacity of the differential data. When driving a driving route having many commonalities, it is expected that the information input to the automatic driving control unit during driving and the information output from the automatic driving control unit will be similar to the information contained in the log data LOG stored in the past. In other words, the capacity of the differential data can be reduced, and the remaining capacity of the storage device 120 can be effectively preserved. In this way, according to the automatic driving system 100 according to the present embodiment, it is possible to effectively prevent the remaining capacity of the storage device 120 from becoming tight while securing the information necessary to ensure explainability and verifiability.

[0059] 5. Variations 5-1. The first variant 5 is a flowchart showing an example of a process for generating a driving route in the first modified example. The timing of execution of the process is similar to that of the flowchart in FIG.

[0060] The processing from step S310 to step S340 is the same as the processing from step S110 to step S140 in Fig. 3. However, in step S330, if the remaining capacity of storage device 120 is equal to or greater than the first predetermined amount (step S330; No), the processing proceeds to step S350. Note that the first predetermined amount is the same as the predetermined amount in step S130 in Fig. 3.

[0061] In step S350, processor 110 determines whether the remaining capacity of storage device 120 is greater than a second predetermined amount. The second predetermined amount may be greater than or equal to the first predetermined amount. If the remaining capacity of storage device 120 is equal to or less than the second predetermined amount (step S350; No), processing proceeds to step S360. If the remaining capacity of storage device 120 is greater than the second predetermined amount (step S350; Yes), processing proceeds to step S370.

[0062] The process of step S360 is the same as the process of step S160 in Fig. 3. On the other hand, in step S370, processor 110 selects a travel route from among the candidate routes using the degree of commonality with travel history DAT as an index. Specifically, processor 110 selects a candidate route that has few commonalities with travel history DAT as the travel route. When the travel route is selected in step S370, the series of processes ends.

[0063] The control data acquired during the automatic driving and stored in the storage device 120 can be used as log data LOG during the subsequent driving of the vehicle 1. It can also be used as learning data for a machine learning model for the subsequent automatic driving control. Therefore, in this modification, when the remaining capacity of the storage device 120 is greater than a predetermined amount, a driving route is selected that is expected to acquire a large amount of control data that is a difference from data acquired in the past. In other words, when the remaining capacity of the storage device 120 is sufficient, the automatic driving system 100 selects a driving route that can collect a large amount of new data that has not been experienced in the past.

[0064] The index for determining the number of commonalities is basically the reverse of the index for determining the number of commonalities described in step S140 of the flowchart in Fig. 3. That is, in step S370, a candidate route that has few commonalities with the travel history DAT is selected based on the number of commonalities of the routes, the number of commonalities of the lanes on which the vehicle 1 travels, the number of commonalities of the road shapes, and the like.

[0065] For example, the processor 110 may select a candidate route that passes through a route that includes many road shapes that are not included in the driving history DAT, or may select a candidate route that travels in different lanes on the same route (for example, a candidate route that travels in the right lane when the driving history DAT includes data of traveling in the left lane).

[0066] Alternatively, from the viewpoint of utilization as a new driving history or learning data, the criteria for determining the lack of commonalities with the driving history DAT may further include viewpoints such as time period and frequency. For example, a driving route that includes many routes that have been traveled infrequently in the past may be determined to have few commonalities. Alternatively, even if a driving route passes through the same route, if the route is traveled during a time period that has not been traveled in the past, it may be determined to have few commonalities. In this way, a candidate route that includes many elements that the vehicle 1 has not experienced in the past is determined to have few commonalities.

[0067] 5-2. The second variant Next, a second modified example will be described. In the second modified example, the vehicle 1 is connected to an external management server via a communication network. In this case, a part of the processor 110 and the storage device 120 may be included in the management server, and at least a part of the processing shown in the flowchart of FIG. 3 may be performed in the management server. In this case, for example, at the timing of determining the travel route, the starting point, the destination, and the remaining capacity of the storage device 120 are transmitted from the vehicle 1 to the management server. Then, the processor of the management server executes the processing shown in the flowchart of FIG. 3 based on the received information and the information held by the management server, and determines the travel route. Then, the determined travel route is transmitted to the vehicle 1. Alternatively, the generation of the candidate route may be performed in the management server, and the selection of the travel route may be performed in the vehicle. In this case, the starting point and the destination are transmitted to the management server, and the management server generates a candidate route based on this information and its own information, and transmits the generated one or more candidate routes to the vehicle 1.

[0068] The management server may also be a server that manages multiple vehicles 1. In this case, the driving history DAT may be data that is an aggregate of data obtained from multiple vehicles 1. In this case, the processor 110 and a part of the storage device 120 may also be included in the management server. In other words, the processing entity that generates candidate routes and selects driving routes may be the management server. In addition, a part or all of the driving history DAT may be stored in the storage device of the management server, and the management server may distribute the stored data to the vehicles 1.

[0069] When the driving history DAT is data obtained from multiple vehicles 1, the commonalities between the driving history DAT and the candidate route do not necessarily mean commonalities with the routes, lanes, or road shapes that the vehicle 1 has experienced in the past. In other words, a candidate route that has many commonalities with routes that a vehicle 1 other than the vehicle 1 has traveled in the past may be selected as the driving route. By aggregating data from multiple vehicles 1, the driving history DAT can be collected more efficiently, making it possible to reduce the differential data.

[0070] The first and second modified examples can be combined. When the process shown in the first modified example is performed as the process for generating a driving route, and when a plurality of vehicles 1 are managed by the management server, the following embodiment is also possible. When there is insufficient data in the driving history DAT aggregated by the management server, the management server prompts the user of the vehicle 1 to drive a driving route that allows the data to be collected. For example, assume that the driving history DAT does not include data on the vehicle 1 driving a certain route. In this case, if there is a margin in the remaining capacity of the storage device 120, it is desirable to select a candidate route that includes the route as the driving route, but this may result in a longer driving time. In such a case, when a candidate route that has few commonalities with the driving history DAT is selected, the management server provides an incentive, such as a point return, to the user, so that a driving route that allows the user to obtain a lot of new data while maintaining the user's satisfaction is selected. [Explanation of symbols]

[0071] 1...vehicle, 10...sensor group, 20...recognition unit, 30...planning unit, 40...control amount calculation unit, 50...travel route management unit, 60...travel device, 100...automatic driving system, 110...processor, 120...storage device, DAT...travel history, CON...control amount, RES...recognition result information, SEN...sensor detection information, TRJ...target trajectory

Claims

1. A driving route management system that generates a driving route in automatic driving control of a vehicle, A storage device that stores a travel history that manages the routes that the vehicle has traveled under automatic driving control in the past; one or more processors; Equipped with The vehicle includes a target storage device that stores log data related to autonomous driving control, The one or more processors: A process of acquiring a remaining capacity of the target storage device; generating one or more candidate routes; selecting the travel route from the one or more candidate routes; configured to run The process of selecting the travel route includes, when the remaining capacity is less than a predetermined amount, selecting the travel route from the one or more candidate routes using a number of commonalities with the travel history as an index. A driving route management system characterized by the above.

2. The route management system according to claim 1, The process of selecting the travel route includes, when the remaining capacity is greater than a predetermined amount, selecting the travel route from among the one or more candidate routes using a small number of commonalities with the travel history as an index. A driving route management system characterized by the above.

3. The route management system according to claim 1 or 2, The driving history includes data regarding routes traveled by the vehicle in the past; The one or more processors: Among the one or more candidate routes, a candidate route that includes a greater number of routes that the vehicle has traveled in the past is determined to have a greater number of common points. A route management system configured as follows.

4. The route management system according to claim 3, The driving history further includes data regarding lanes on which the vehicle has traveled in the past; The one or more candidate routes further include information specifying lanes to be traveled in the route; The one or more processors: A candidate route that includes more lanes on which the vehicle has traveled in the past among the one or more candidate routes is determined to have more common points. A route management system configured as follows.

5. The route management system according to claim 1 or 2, The driving history includes information regarding a specific road shape existing on a route traveled by the vehicle in the past, The one or more processors: further performing a process of acquiring information on road shapes included in each of the one or more candidate routes; a candidate route including a greater number of road shapes that are the same as the specific road shape among the one or more candidate routes is determined to have a greater number of common points; A route management system configured as follows.

Images