A kind of automatic driving domain controller and vehicle
By using multiple controllers to divide the work into the autonomous driving domain controller, the problem of insufficient processing capabilities in the existing technology is solved, and the real-time and safety of autonomous driving is improved.
Patent Information
- Application Number
- CN202111473275.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2021-12-02
- Publication Date
- 2024-11-05
- Estimated Expiration
- 2041-12-02
AI Technical Summary
The insufficient processing capabilities of existing autonomous driving domain controllers limit the improvement of real-time and safety of autonomous driving, especially with the increase in sensor types and data volume.
Multiple controllers are used to process data together, distribute processing resources, perform vehicle control and data processing through the first controller, and monitor failures of the first controller through the second controller to improve the security and real-timeness of the system.
Through the division of labor of multiple controllers, the limitation of real-time and security of insufficient processor processing capabilities is reduced, and the real-time and security of autonomous driving is improved. In particular, the monitoring function of the second controller is further improved system stability. sex.
Smart Images

Figure CN114237104B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of vehicles, and in particular to an autonomous driving domain controller and a vehicle. Background Art
[0002] As vehicles become more intelligent, autonomous driving technology has come into people’s view. Autonomous driving technology can automate the work performed by the driver by automatically controlling the vehicle’s operating status.
[0003] In order to achieve vehicle control during the autonomous driving process, sensors are needed to collect data on the vehicle and the vehicle's environment, and process the data to obtain control results. The entire process depends on the autonomous driving domain controller.
[0004] The autonomous driving domain controller can be regarded as the brain that controls the vehicle to perform autonomous driving. It is usually connected to multiple sensors and has functions such as multi-sensor fusion, positioning, path planning, decision-making, and control. As the requirements for the real-time and safety of autonomous driving increase, the types of sensors connected to the autonomous driving domain controller are increasing, and the amount of data collected is increasing. Therefore, the processor capacity required to process the increasing amount of data has increased.
[0005] However, the current processing power of autonomous driving domain controllers limits the improvement of autonomous driving real-time and safety. Summary of the invention
[0006] In view of this, the present application provides an autonomous driving domain controller and a vehicle to improve the real-time and safety of autonomous driving.
[0007] In a first aspect, the present application provides an autonomous driving domain controller, the autonomous driving domain controller comprising a first controller and a second controller, wherein:
[0008] The first controller is connected to the second controller;
[0009] The first controller is used to process data collected by sensors in the first sensor set to obtain first data;
[0010] The second controller is used to process the data collected by the sensors in the second sensor set to obtain second data, send the second data to the first controller, and send a fault instruction to the first controller when a fault occurs in the first controller; the processing resources required to obtain the second data are greater than the processing resources required to obtain the first data;
[0011] The processing resources required to obtain the second data are greater than the processing resources required to obtain the first data, that is, the processing resources required for processing the data collected by the sensors in the second sensor set to obtain the second data and processing the data collected by the sensors in the first sensor set to obtain the first data are greater than the processing resources required for the latter processing process.
[0012] The first controller is also used to control the vehicle according to the first data and the second data, and enter an automatic driving fault state when receiving the fault instruction.
[0013] Compared with one controller completing all processing steps during the autonomous driving process, the solution of this embodiment can complete the processing steps through multiple controllers, reducing the limitations on the real-time and safety of autonomous driving caused by insufficient processor processing power; moreover, the first controller controls the vehicle, that is, the first controller simultaneously performs the process of obtaining the first data and the process of controlling the vehicle, while the second controller performs the process of obtaining the second data. Since the second controller does not need to directly control the vehicle under normal circumstances, more processing resources can be allocated to the process of obtaining the second data. Therefore, the second controller performs the processing process requiring more processing resources, that is, the processing of obtaining the second data. Compared with both controllers directly controlling the vehicle, the solution of this application can improve the real-time performance of autonomous driving; in addition, the second controller monitors the processing process of the first controller to further improve the safety of autonomous driving.
[0014] In a possible implementation, the autonomous driving domain controller further includes a third controller, wherein:
[0015] The third controller is connected to the first controller and the second controller respectively;
[0016] The third controller is used to send a second fault instruction to the first controller when monitoring that the second controller fails;
[0017] The first controller is also used to enter an automatic driving fault state when receiving the second fault instruction.
[0018] In a possible implementation manner, the first controller includes a first core and a second core, wherein:
[0019] The first core is used to process data collected by sensors in the first sensor set to obtain first data;
[0020] The second core is used to control the vehicle according to the first data and the second data, and enter an automatic driving fault state when receiving the fault instruction;
[0021] The data processing efficiency of the first core is higher than the data processing efficiency of the second core.
[0022] In a possible implementation manner, the second controller includes a third core and a fourth core, wherein:
[0023] The third core is used for processing the data collected by the sensors in the second sensor set to obtain second data;
[0024] The fourth core is configured to send a fault instruction to the first controller when a fault is detected in the first controller;
[0025] The data processing efficiency of the third core is higher than the data processing efficiency of the fourth core.
[0026] In a possible implementation manner, the first controller is further configured to send status information to the second controller;
[0027] The second controller is further used to determine whether a fault occurs in the first controller according to the received status information.
[0028] In a possible implementation manner, the second controller is further configured to send second status information to the third controller;
[0029] The third controller is further used to determine whether the second controller fails according to the received second status information.
[0030] In a possible implementation manner, the sensors in the second sensor set include a surround view camera and an inertial sensor;
[0031] The second controller is specifically used to process the data collected by the surround view camera, and to perform dead reckoning based on the data collected by the inertial sensor to obtain the second data; the dead reckoning refers to using the known position of the vehicle at a certain moment and the subsequent displacement to estimate the vehicle's subsequent position. In one possible implementation, the first controller is specifically used to disable the automatic driving control function when the fault instruction is received. In one possible implementation, the first controller is specifically used to determine the vehicle's driving path based on the first data and the second data;
[0032] The vehicle is controlled according to the vehicle travel path.
[0033] In a second aspect, the present application provides a vehicle equipped with any of the above-mentioned autonomous driving domain controllers. BRIEF DESCRIPTION OF THE DRAWINGS
[0034] Figure 1 It is a schematic diagram of the structure of the autonomous driving domain controller provided in an embodiment of the present application;
[0035] Figure 2 It is a structural diagram of an autonomous driving domain controller provided in another embodiment of the present application. DETAILED DESCRIPTION
[0036] As the requirements for real-time and safety of autonomous driving increase, the types of sensors connected to autonomous driving domain controllers are increasing, the amount of data collected is increasing, and the processor capacity required to process the increasing amount of data is increasing. However, the current processing capacity of autonomous driving domain controllers limits the improvement of real-time and safety of autonomous driving.
[0037] Based on this, in the embodiment of the present application provided by the inventor, the autonomous driving domain controller includes a first controller and a second controller. The first controller and the second controller are connected; the first controller is used to process the data collected by the sensors in the first sensor set to obtain the first data; the second controller is used to process the data collected by the sensors in the second sensor set to obtain the second data, send the second data to the first controller, and when the first controller is monitored to have a fault, send a fault instruction to the first controller; the processing resources required to obtain the second data are greater than the processing resources required to obtain the first data; the first controller is also used to control the vehicle according to the first data and the second data, and enter the autonomous driving fault state when receiving the fault instruction.
[0038] Compared with one controller completing all processing steps during the autonomous driving process, the solution of this embodiment can be used to complete the processing steps through multiple controllers, thereby reducing the limitations on the real-time and safety of autonomous driving caused by insufficient processor processing power; moreover, the first controller controls the vehicle, that is, the first controller simultaneously performs the process of obtaining the first data and the process of controlling the vehicle, while the second controller performs the process of obtaining the second data. Since the second controller does not need to directly control the vehicle under normal circumstances, more processing resources can be allocated to the process of obtaining the second data. Therefore, the second controller performs the processing process requiring more processing resources, that is, performs the process of obtaining the second data. Compared with both controllers directly controlling the vehicle, the solution of this application can improve the real-time performance of autonomous driving; in addition, the second controller monitors the first controller, which can further improve the safety of autonomous driving.
[0039] In order to facilitate understanding of the technical solution provided in the embodiment of the present application, an autonomous driving domain controller and a vehicle provided in the embodiment of the present application are described below with reference to the accompanying drawings.
[0040] Although the exemplary embodiments of the present application are shown in the drawings, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments described herein. Based on the embodiments in the present application, other embodiments obtained by those skilled in the art without making creative contributions are all within the scope of protection of the present application.
[0041] In the claims and specification and drawings of this application, the terms "includes" and "has" and any variations thereof are intended to cover non-exclusive inclusions.
[0042] This application provides an autonomous driving domain controller. Figure 1 , Figure 1 It is a structural diagram of the autonomous driving domain controller provided in an embodiment of the present application.
[0043] like Figure 1 As shown, the autonomous driving domain controller 100 in the embodiment of the present application includes a first controller 101 and a second controller 102.
[0044] The first controller is connected to the second controller 102 .
[0045] The first controller 101 is used to process data collected by sensors in the first sensor set 103 to obtain first data.
[0046] The second controller 102 is used to process the data collected by the sensors in the second sensor set 104 to obtain second data.
[0047] In order to achieve autonomous driving, multiple / various sensors are usually required to collect data and process the data collected by the sensors.
[0048] The above-mentioned sensor set, ie, the sensor set and the second sensor set, refers to a set including a plurality of sensors, ie, the above-mentioned sensor set includes a plurality of sensors.
[0049] The sensors in the first sensor set and the second sensor set may be of the same type or of different types.
[0050] The first data mentioned above refers to data obtained by the first controller 101 through a process such as processing.
[0051] The second data mentioned above refers to data obtained by the second controller 102 through a process such as processing.
[0052] The processing resources required to obtain the second data are greater than the processing resources required to obtain the first data.
[0053] That is, the data collected by the sensors in the second sensor set are processed to obtain the second data, and the data collected by the sensors in the first sensor set are processed to obtain the first data. For the processing resources required for these two processing processes, the latter is greater than the former.
[0054] The second controller 102 is further configured to send second data to the first controller 101 , and send a fault instruction to the first controller 101 when it is monitored that a fault occurs in the first controller 101 .
[0055] The first controller 101 is also used to control the vehicle according to the first data and the second data, and enter the automatic driving fault state when a fault instruction is received.
[0056] The second controller 102 monitors the first controller, and when a fault is detected, a fault instruction is sent to the first controller, so that the first controller 101 enters an automatic driving fault state, so as to improve the safety of automatic driving.
[0057] Compared with one controller completing all processing steps during the autonomous driving process, the solution of this embodiment can complete the processing steps through multiple controllers, reducing the limitations on the real-time and safety of autonomous driving caused by insufficient processor processing power; moreover, the first controller controls the vehicle, that is, the first controller simultaneously performs the process of obtaining the first data and the process of controlling the vehicle, while the second controller performs the process of obtaining the second data. Since the second controller does not need to directly control the vehicle under normal circumstances, more processing resources can be allocated to the process of obtaining the second data. Therefore, the second controller performs the processing process requiring more processing resources, that is, the processing of obtaining the second data. Compared with both controllers directly controlling the vehicle, the solution of this application can improve the real-time performance of autonomous driving; in addition, the second controller monitors the processing process of the first controller to further improve the safety of autonomous driving.
[0058] The following is an explanation in conjunction with a specific implementation method.
[0059] This application also provides another autonomous driving domain controller. Figure 2 , Figure 2 It is a structural diagram of an autonomous driving domain controller provided in another embodiment of the present application.
[0060] like Figure 2 As shown, the autonomous driving domain controller 100 in the embodiment of the present application includes: a first controller 101, a second controller 102 and a third controller 105.
[0061] The first controller 101 , the second controller 102 and the third controller 105 are connected in pairs, that is, the first controller 101 is connected to the second controller 102 , the first controller 101 is connected to the third controller 105 , and the second controller 102 is connected to the third controller 105 .
[0062] The first controller 101 includes a first core 1011 and a second core 1012 , and the second core 1012 includes a first monitoring module 1013 and a control module 1014 ;
[0063] The second controller 102 includes a third core 1021 and a fourth core 1022 , and the fourth core 1022 includes a second monitoring module 1023 ;
[0064] The third controller 105 includes a third monitoring module 1051;
[0065] The first core 1011 is connected to sensors in the first sensor set 103 , and the third core 1021 is connected to sensors in the second sensor set 104 .
[0066] During the autonomous driving process, path planning and controlling the vehicle status based on the planning results are important application scenarios of the domain controller.
[0067] The following uses the path planning scenario as an application scenario to explain the autonomous driving domain controller in the embodiment of the present application.
[0068] The sensors in the first sensor set 103 include: a radar, a front-view camera, and a rear-view camera.
[0069] The first core 1011 is connected to the sensors in the first sensor set 103 and is used to receive and process data collected by the sensors in the first sensor set 103 to obtain first data.
[0070] Regarding the data processing process, the first core 1011 is specifically used for: radar perception processing, front camera perception processing, rear camera perception processing, multi-sensor fusion, and path planning based on the results of multi-sensor fusion.
[0071] Multi-sensor Fusion (MSF) is an information processing process that uses computer technology to analyze and process data from multiple sensors or multiple sources according to preset criteria to complete the required decision-making and estimation. The data used for multi-sensor fusion is usually collected by multiple or multiple sensors.
[0072] Through multi-sensor fusion, the data collected by multiple / various sensors are comprehensively utilized to analyze vehicle information and the vehicle's environmental information, making the path planning results more scientific.
[0073] In a possible manner, the first core 1011 may include a first receiving module and a first processing module, which are respectively used to receive and process data collected by sensors in the first sensor set 103 .
[0074] In some possible situations, the processing of data collected by different sensors (radar perception processing, front-view camera perception processing, rear-view camera perception processing) can be completed through different processing modules.
[0075] In some possible cases, the modules used for radar perception processing, front-view camera perception processing, and rear-view camera perception processing can be distinguished from the modules used for multi-sensor fusion and path planning.
[0076] In a possible manner, the first core 1011 may also be used to control sensors in the first sensor set 103 to collect data.
[0077] Furthermore, in some possible cases, the first core 1011 further includes a first control module, which is used to control the sensors in the first sensor set 103 to collect data.
[0078] The control module 1014 is used to control the vehicle according to the path planning result obtained by the first core 1011 .
[0079] Specifically, the control module 1014 controls the vehicle, including controlling the operating state of the vehicle, for example, controlling various execution units such as the steering system, braking system, and power system of the vehicle.
[0080] The first monitoring module 1013 is used to perform first level monitoring, specifically including: monitoring the first core 1011 .
[0081] Specifically, the first monitoring module 1013 is used to monitor: radar perception processing, front camera perception processing, rear camera perception processing, multi-sensor fusion, and path planning.
[0082] In some possible situations, the first monitoring module 1013 may also be used to monitor the first core 1011 receiving data.
[0083] In some possible situations, the monitoring content may include: data transmission delay and program execution status, such as program failure to execute, incomplete execution, and execution errors.
[0084] The first monitoring module 1013 is used to perform the first level monitoring, and when a fault occurs, sends a blocking instruction to the control module 1014 to make the first controller 101, for example, make the control module 1014 enter the automatic driving fault state.
[0085] In one possible implementation, the autonomous driving failure state may include: disabling the autonomous driving function, or entering a pre-set safe state.
[0086] The first-level monitoring function implemented by the first monitoring module 1013 can monitor the processing process during the automatic driving process, and enter the automatic driving fault state when a fault is found, thereby improving the safety of automatic driving.
[0087] The first controller 101 includes multiple cores. The first core 1011 processes the data collected by the sensor, and the second core 1012 performs monitoring, and can reasonably allocate processing resources according to actual needs, so as to reduce the cost of the domain controller while improving the safety and real-time performance of the autonomous driving.
[0088] Furthermore, in some possible implementations, the data processing efficiency of the first core 1011 is higher than the data processing efficiency of the second core 1012 .
[0089] The first core 1011 is used to receive and process data collected by sensors in the first sensor set 103 to obtain first data; the second core 1012 includes a first monitoring module 1013 and a control module 1014, which are used for monitoring and control.
[0090] Since more processing resources are needed to process the data obtained by the sensors during the process of autonomous driving, the first core 1011 has a higher data processing efficiency and can process the data collected by the detector faster, thereby improving the real-time performance of autonomous driving; in addition, the data processing efficiency of the first core 1011 is higher than that of the second core 1012, which can reduce the cost of the autonomous driving domain controller on the basis of improving the real-time performance of autonomous driving, thereby reducing the cost of the entire vehicle.
[0091] Furthermore, in some possible implementations, the second core 1012 has high functional stability (or functional safety), that is, data processing stability. The stability here means that after the second core 1012 completes multiple data processing, it can better maintain its own functions. For example, after completing multiple processing, the second core 1012 rarely has random failures and the like.
[0092] Since the second core 1012 is used to implement control functions and monitoring functions, it has high functional stability and can more stably implement vehicle control and monitoring of the controller, reducing errors in the control and monitoring process, thereby improving the safety of autonomous driving.
[0093] Furthermore, in some possible situations, in order to reduce the cost of the autonomous driving domain controller, the functional stability of the second core 1012 may be higher than that of the first core 1011 .
[0094] The sensors in the second sensor set 104 include: a surround view camera and an inertial measurement unit (IMU).
[0095] The third core 1021 is connected to the sensors in the second sensor set 104, and is used for receiving and processing the data collected by the sensors in the second sensor set 104 to obtain second data; and is also used for sending the second data to the first controller.
[0096] Regarding the data processing process, the third core 1021 is specifically used for: surround view camera perception processing and dead reckoning based on inertial sensor perception processing.
[0097] The processing resources required to obtain the second data are greater than the processing resources required to obtain the first data.
[0098] In order to realize the automatic driving of the vehicle, it is usually necessary to collect data from multiple detectors and process the data according to the processing results. Generally, among the multiple detectors, compared with radar perception processing, front camera perception processing, rear camera perception processing, multi-sensor fusion, and path planning based on the results of multi-sensor fusion, surround view cameras and dead reckoning based on the data collected by the inertial sensors require larger processing resources.
[0099] Other possible implementations and possible situations of the third core 1021 are similar to those of the first core 1011 and will not be described in detail here.
[0100] In this embodiment, the third core 1021 processes the data collected by the sensors in the second sensor set 104 .
[0101] Compared with completing the above processing by one controller, the solution of this embodiment can be used to complete the processing through multiple controllers, thereby reducing the situation where insufficient processor computing power limits the safety and real-time performance of autonomous driving.
[0102] Under normal circumstances, the second controller does not need to directly control the vehicle, and can allocate more processing resources to the process of obtaining the second data. Therefore, the second controller performs the process requiring more processing resources, that is, the process of obtaining the second data. Compared with both controllers directly controlling the vehicle, the solution of the present application can improve the real-time performance of the autonomous driving.
[0103] The third core 1021 is further configured to send the second data to the first core 1011 , so that the first core 1011 performs multi-sensor fusion and path planning according to the second data.
[0104] In a possible implementation, the first core 1011 performs multi-sensor fusion and path planning according to the second data, which may include: performing multi-sensor fusion and path planning according to the first data and the second data.
[0105] The second monitoring module 1023 is used to perform second-level monitoring, specifically including: monitoring the second core 1012 and the third core 1021 .
[0106] The second core 1012 is further configured to send status information to the second monitoring module 1023 .
[0107] The second monitoring module 1023 is specifically configured to determine whether a fault occurs in the second core 1012 according to the received status information.
[0108] The status information refers to the status of the first controller, and may specifically include status information of the second core.
[0109] Specifically, the second monitoring module 1023 is used to monitor the second core 1012, including monitoring the first monitoring module 1013 and the control module 1014; the second monitoring module 1023 is used to monitor the third core 1021, including monitoring: surround view camera perception processing, dead reckoning based on inertial sensor perception processing.
[0110] In some possible situations, the second monitoring module 1023 may also be used to monitor the third core 1021 receiving data.
[0111] The second monitoring module 1023 is used to perform the second level monitoring, and when a fault occurs, sends a blocking instruction to the control module 1014 to make the first controller 101, such as the control module 1014, enter the automatic driving fault state.
[0112] The second monitoring module 1023 is also used to control the vehicle when it is monitored that a fault occurs in the first controller.
[0113] When the first controller fails, controlling the vehicle by the first controller may be dangerous. In this case, the second controller may be used to control the vehicle.
[0114] When the second monitoring module 1023 is used for the second level monitoring, the specific content, specific method, and the effect that can be achieved are similar to those of the first monitoring module 1013. Therefore, they will not be described in detail here.
[0115] The second controller 1021 includes multiple cores. The third core 1021 completes the processing of the data collected by the sensor, and the fourth core 1022 completes the monitoring, and can reasonably allocate processing resources according to actual needs to reduce the cost of the domain controller while improving the safety and real-time performance of autonomous driving.
[0116] Furthermore, in some possible implementations, the data processing efficiency of the third core 1021 is higher than the data processing efficiency of the fourth core 1022 .
[0117] The third core 1021 is used to receive and process data collected by sensors in the second sensor set 104 to obtain second data; the fourth core 1022 includes a second monitoring module 1023 for monitoring.
[0118] Since more processing resources are needed to process the data obtained by the sensors during the process of autonomous driving, the third core 1021 has a higher data processing efficiency and can process the data collected by the detector faster, thereby improving the real-time performance of autonomous driving; in addition, the data processing efficiency of the third core 1021 is higher than that of the fourth core 1022, which can reduce the cost of the domain controller on the basis of improving the real-time performance of autonomous driving, thereby reducing the cost of the entire vehicle.
[0119] Furthermore, in some possible implementations, the fourth core 1022 has high functional stability (or functional safety), that is, data processing stability. The stability here means that after the fourth core 1022 completes multiple data processing, it can better maintain its own functions. For example, after completing multiple processing, the fourth core 1022 rarely has random failures and the like.
[0120] Since the fourth core 1022 is used to implement control functions and monitoring functions, it has high functional stability and can more stably implement vehicle control and monitoring of the controller, reducing errors in the control and monitoring process, thereby improving the safety of autonomous driving.
[0121] Further, in some possible cases, in order to reduce the cost of the autonomous driving domain controller, the functional stability of the fourth core 1022 may be higher than the functional stability of the third core 1012 .
[0122] The third monitoring module 1051 is used to perform third-level monitoring, specifically including: monitoring the fourth core 1022 .
[0123] Specifically, the third monitoring module 1051 is used to monitor the second monitoring module 1023 .
[0124] The second monitoring module 1023 is further configured to send second status information to the third monitoring module 1051 .
[0125] The third monitoring module 1051 is specifically configured to determine whether the fourth core 1022 fails according to the received second status information.
[0126] The second status information refers to the status of the second controller, and may specifically include status information of the fourth core 1022 and / or status information of the second monitoring module 1023 .
[0127] The third monitoring module 1051 is used to perform the third level monitoring. When a fault occurs, it sends a blocking instruction to the control module 1014 to make the first controller 101, such as the control module 1014, enter the automatic driving fault state.
[0128] The third monitoring module 1051 is also used to control the vehicle when it is monitored that a fault occurs in the second controller.
[0129] When the second controller fails, the vehicle may be controlled by the first controller / second controller, which may be dangerous. In this case, the vehicle may be controlled by the third controller.
[0130] When the third monitoring module 1051 is used for the third level monitoring, the specific content, specific method, and the effect that can be achieved are similar to those of the first monitoring module 1013. Therefore, they will not be described in detail here.
[0131] The first monitoring module 1013, the second monitoring module 1023 and the third monitoring module 1051 are used for first-level monitoring, second-level monitoring and third-level monitoring, respectively. Although the present embodiment adopts a step-by-step monitoring strategy, when a failure occurs in any of the three levels, a blocking instruction is sent to the control module 1014 to make the first controller 101 enter an automatic driving failure state, thereby improving the safety of automatic driving.
[0132] For example, when the third monitoring module 1051 monitors that the second controller 102 has a fault, the second monitoring module 1023's monitoring of the first controller 101 is unreliable. Therefore, a fault instruction is directly sent to the control module 1014 of the first controller 101 through the third monitoring module 1051 to make the first controller 101 enter an automatic driving fault state, thereby reducing safety issues caused by unreliable monitoring results of the second monitoring module 1023, thereby improving the safety of automatic driving.
[0133] The autonomous driving domain controller in this embodiment adopts a three-level monitoring strategy.
[0134] Compared with single-level monitoring, three-level monitoring is adopted in this embodiment, which can achieve a more stable monitoring effect through a hierarchical monitoring strategy; moreover, since the third controller is not used to process the data collected by the sensor, the third controller is less likely to have failures caused by processing large amounts of data than the first controller and the second controller, and can achieve more stable monitoring; in addition, the third-level monitoring can identify the failure of the second controller, and by improving the stability of the second-level monitoring, the stability of the monitoring of the first controller is improved, thereby improving the safety of autonomous driving; furthermore, when a failure occurs at any level of the three-level monitoring, a blocking instruction will be sent to cause the first controller to enter an autonomous driving failure state, thereby reducing safety issues caused by unreliable monitoring results at the intermediate level, thereby improving the safety of autonomous driving.
[0135] In summary, the autonomous driving domain controller with a three-level monitoring strategy can improve the safety of autonomous driving.
[0136] The first controller 101 includes a first core 1011 and a second core 1012 , and the second controller 102 includes a third core 1021 and a fourth core 1022 .
[0137] Since the first core 1011 and the third core 1021 are used to perform a large amount of data processing, the first core 1011 and the third core 1021 may include a controller with strong computing capabilities.
[0138] Since the second core 1012 is used for monitoring and controlling the vehicle, and the fourth core 1022 is used for monitoring, the second core 1012 and the fourth core 1022 may include a controller with strong functional stability to improve the stability of the control process and the monitoring process, thereby improving the safety of autonomous driving.
[0139] The chip model TDA4 may include multiple cores. For example, a TDA4 chip includes two cores, namely, an A core with stronger computing power and an M core with stronger functional stability.
[0140] Specifically, the TDA4 chip has passed the ISO26262 functional safety certification standard, wherein the functional safety level of the A core of the TDA4 chip is ASIL-B, and the functional safety level of the M core of the TDA4 chip is ASIL-D, that is, the M core of the TDA4 chip has a functional safety level two levels higher than that of the A core of the TDA4 chip.
[0141] In a possible implementation, based on the characteristics of the TDA4 chip, both the first controller 101 and the second controller 102 may use the TDA4 chip.
[0142] Specifically, the first controller 101 is taken as an example for description here, and the second controller 102 is similar to the first controller 101 and will not be described in detail here.
[0143] The first controller 101 adopts a TDA4 chip, with the A core of the chip being used as the first core 1011 of the first controller 101, and the A core in the TDA4 chip having higher processing efficiency is used to process data; the M core of the chip is used as the second core 1012 of the first controller 101, and the M core in the TDA4 chip having higher functional stability is used to implement monitoring and control processes, thereby better exerting the functions of the chip.
[0144] Similarly, the second controller 102 also uses a TDA4 model chip, with the A core of the chip serving as the second core 1021 of the second controller 102, and the M core of the chip serving as the fourth core 1022 of the second controller 102. The reasons for the selection are similar to those for the first controller 101, which will not be repeated here.
[0145] The third controller 103 is used to monitor the processing process and results of the fourth core 1022 in the second controller 102. The specific chip model can be determined according to actual needs. For example, considering the cost and technical maturity, a chip model of TC397 can be used.
[0146] The following is an example of implementing three-level monitoring using the three selected chips mentioned above. The first controller 101 is TDA4-1 (including A core and M core), the second controller 102 is TDA4-2 (including A core and M core), and the third controller 103 is TC397.
[0147] When the M core of TDA4-2 detects a fault in TDA4-1 (A core or M core), a diagnostic error code (DTC) is generated to prohibit the M core of TDA4-1 from continuing to control the vehicle operation status;
[0148] When TC397 detects a fault in TDA4-2 (A core or M core), it can also generate a diagnostic error code (DTC) to prohibit TDA4-1's M core from continuing to control the vehicle's operating status. At this time, since the monitoring result of TDA4-2 on TDA-1 is unreliable, even if TDA-1 itself does not report an error, TC397 needs to issue a command to block TDA-1's M core from controlling the vehicle's operating status to improve the safety of autonomous driving.
[0149] The present application also provides a vehicle.
[0150] The vehicle in the embodiments of the present application is equipped with any of the autonomous driving domain controllers in the above embodiments.
[0151] The autonomous driving domain controller included in the above-mentioned vehicle can achieve the same technical effect as the autonomous driving domain controller in the above embodiments. To avoid repetition, it will not be repeated here.
[0152] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.
Claims
1. An autonomous driving domain controller, characterized in that: The autonomous driving domain controller includes a first controller, a second controller and a third controller, wherein: The first controller is connected to the second controller; The first controller is used to process data collected by sensors in the first sensor set to obtain first data; The second controller is used to process the data collected by the sensors in the second sensor set to obtain second data, send the second data to the first controller, and send a fault instruction to the first controller when it is monitored that the first controller fails; the processing resources required to obtain the second data are greater than the processing resources required to obtain the first data; The first controller is further configured to control the vehicle according to the first data and the second data, and enter an automatic driving fault state when receiving the fault instruction; The second controller is further used to control the vehicle when the first controller fails; the second controller does not directly control the vehicle when the first controller does not fail; The third controller is connected to the first controller and the second controller respectively; the third controller is not used to process the data collected by the sensor; The third controller is used to send a second fault instruction to the first controller when monitoring that the second controller fails; The first controller is also used to enter an automatic driving fault state when receiving the second fault instruction.
2. The autonomous driving domain controller according to claim 1, characterized in that: The first controller includes a first core and a second core, wherein: The first core is used to process data collected by sensors in the first sensor set to obtain first data; The second core is used to control the vehicle according to the first data and the second data, and enter an automatic driving fault state when receiving the fault instruction; The data processing efficiency of the first core is higher than the data processing efficiency of the second core.
3. The autonomous driving domain controller according to claim 1, characterized in that: The second controller includes a third core and a fourth core, wherein: The third core is used for processing the data collected by the sensors in the second sensor set to obtain second data; The fourth core is configured to send a fault instruction to the first controller when a fault is detected in the first controller; The data processing efficiency of the third core is higher than the data processing efficiency of the fourth core.
4. The autonomous driving domain controller according to claim 1, characterized in that: The first controller is further configured to send status information to the second controller; The second controller is further used to determine whether a fault occurs in the first controller according to the received status information.
5. The autonomous driving domain controller according to claim 1, characterized in that: The second controller is further used to send second status information to the third controller; The third controller is further used to determine whether the second controller fails according to the received second status information.
6. The autonomous driving domain controller according to claim 1, characterized in that: The sensors in the second sensor set include a surround view camera and an inertial sensor; The second controller is specifically used to process the data collected by the surround view camera and perform dead reckoning based on the data collected by the inertial sensor to obtain the second data.
7. The autonomous driving domain controller according to claim 1, characterized in that: The first controller is specifically used for: When the fault instruction is received, the automatic driving control function is disabled.
8. The autonomous driving domain controller according to claim 1, characterized in that: The first controller is specifically configured to determine a vehicle driving path according to the first data and the second data; The vehicle is controlled according to the vehicle travel path.
9. A vehicle, characterized in that: The vehicle is equipped with an autonomous driving domain controller as described in any one of claims 1 to 8.
