Simulator, vehicle control system, and simulation method

The simulator addresses the lack of comparative analysis in existing safety evaluation devices by generating and analyzing scenarios to evaluate driving support and autonomous driving functions, improving verification efficiency and reducing manual analysis time, thus enhancing the assessment of vehicle control systems.

WO2024241498A1PCT designated stage expired Publication Date: 2024-11-28ASTEMO LTD
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
PCT/JP2023/019170
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Filing Date
2023-05-23
Publication Date
2024-11-28

AI Technical Summary

Technical Problem

Current safety evaluation devices for vehicle driving support and autonomous driving functions lack the ability to perform comparative simulations, leading to increased manual analysis time to assess the performance superiority or inferiority of evaluated functions, and do not provide quality analysis of the driving support and automatic driving functions.

Method used

A simulator that generates a reproduction scenario of a vehicle event, executes simulations using different control parameters, and analyzes the operation quality of driving support and automatic driving functions by comparing the results, thereby improving software verification efficiency.

Benefits of technology

Reduces manual analysis time and enhances verification efficiency by automatically analyzing large-scale simulation results, allowing for the identification of performance issues in driving support and automatic driving functions before mass production and potentially reducing recalls.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure JP2023019170_28112024_PF_FP_ABST
    Figure JP2023019170_28112024_PF_FP_ABST
Patent Text Reader

Abstract

The present invention comprises: a reproduction scenario generation unit for generating a reproduction scenario; an evaluation object simulation execution unit for executing a simulation on the basis of the reproduction scenario by using a first control parameter that operates a driving assistance function and / or an automatic driving function for a vehicle; a comparison simulation execution unit for executing a simulation on the basis of the reproduction scenario using a second control parameter; and an operation quality analysis unit for analyzing the operation quality of the driving assistance function and / or the automatic driving function of the vehicle on the basis of the simulation result outputted from the evaluation object simulation execution unit and the simulation result outputted from the comparison simulation execution unit.
Need to check novelty before this filing date? Find Prior Art

Description

Simulator, vehicle control system, and simulation method

[0001] The present invention relates to a simulator.

[0002] The situations in which driving assistance and autonomous driving functions are actually used in automobiles and other vehicles vary greatly depending on the combination of road shape, driving environment (from expressways with relatively few obstructions to urban areas with many obstructions and pedestrians), weather, etc. In any situation, driving assistance and autonomous driving functions must be developed to ensure safe driving.

[0003] Therefore, in the development of driving assistance and autonomous driving functions, verification technology that utilizes simulations that can reproduce the various scenes mentioned above is important, and technology is being developed to reproduce and verify traffic accidents and near misses in virtual space.

[0004] Background art in this field is found in Patent Document 1 (JP 2008-261991 A), which describes a safety evaluation device including: a simulation means for simulating a vehicle accident by running a vehicle model, which is a model of a vehicle on a virtual road space, with a passenger virtually inside the vehicle model; a prediction means for predicting an injury condition of the passenger based on the occurrence of the vehicle accident simulated by the simulation means; and a display means for displaying the injury condition predicted by the prediction means.

[0005] JP 2008-261991 A

[0006] The safety evaluation device described in Patent Document 1 simulates a vehicle accident by having an accident reproduction unit drive a vehicle model on a virtual road space with a virtual occupant inside, and predicts the injury status of the occupant, but does not evaluate the performance of driving assistance functions or autonomous driving functions.

[0007] Furthermore, the safety evaluation device described in Patent Document 1 executes a simulation only for the function to be evaluated, and does not execute a comparison simulation that is necessary to analyze the quality of the simulation execution result of the function to be evaluated. Therefore, additional man-hours are required to analyze the performance merits or demerits of the function to be evaluated after the simulation is completed.

[0008] Therefore, the present invention aims to provide a simulation device that can automatically perform a good / bad analysis of the performance merits of an application to be evaluated, by performing at least one simulation of a comparison object in addition to the simulation of the application to be evaluated, for driving assistance functions and autonomous driving functions.

[0009] A representative example of the invention disclosed in the present application is as follows: That is, a simulator for evaluating at least one of a driving assistance function and an automatic driving function of a vehicle includes: a reproduction scenario generation unit that reproduces and generates a scenario up to a point where at least one of the driving assistance function and the automatic driving function of the vehicle is activated or a point where an accident involving the vehicle occurs, based on the results of a traffic flow simulation that reproduces traffic phenomena occurring on a road in a virtual road space; an evaluation target simulation execution unit that executes a simulation using first control parameters for activating at least one of the driving assistance function and the automatic driving function of the vehicle, based on the generated reproduction scenario; a comparison simulation execution unit that executes a simulation using second control parameters different from the first control parameters, based on the generated reproduction scenario; and an operation proper / improper analysis unit that analyzes the proper operation of at least one of the driving assistance function and the automatic driving function of the vehicle, based on the simulation results by the evaluation target simulation execution unit and the simulation results by the comparison simulation execution unit.

[0010] According to one aspect of the present invention, it is possible to improve the efficiency of software verification. Problems, configurations, and effects other than those described above will become apparent from the following description of the embodiments.

[0011] FIG. 1 is a block diagram showing an example of the overall configuration of a simulation device according to an embodiment of the present invention. FIG. 2 is a flowchart of processing executed by the simulation device according to an embodiment of the present invention. FIG. 3 is a diagram showing the definition of analysis rules related to a collision scenario according to an embodiment of the present invention. FIG. 4 is a diagram showing the definition of rules for additional analysis when it is determined that "determination is based on the degree of collision" according to an embodiment of the present invention. FIG. 5 is a diagram showing the definition of analysis rules related to a lane keeping scenario according to an embodiment of the present invention. FIG. 6 is a diagram showing the definition of rules for additional analysis when it is determined that "determination is based on the degree of lane departure" according to an embodiment of the present invention. FIG. 7 is a diagram showing the definition of rules for additional analysis when it is determined that "determination is based on the degree of in-lane sway" according to an embodiment of the present invention. FIG. 8 is a diagram showing the definition of analysis rules related to a merging scenario according to an embodiment of the present invention. FIG. 9 is a diagram showing the definition of rules for additional analysis when it is determined that "determination is based on the degree of sudden deceleration" according to an embodiment of the present invention. FIG. 10 is a diagram showing an example of a vehicle control system according to an embodiment of the present invention.

[0012] In an embodiment of the present invention, a simulator 100 is described that uses the results of a large-scale simulation to generate a reproduction scenario that reproduces a scene in which an event (such as the activation of a driving assistance function or an automatic driving function due to an accident or near miss) occurs, and that uses the generated reproduction scenario to compare the results of a simulation performed using first control parameters for at least one of the driving assistance function and the automatic driving function with the results of a simulation performed using second control parameters to analyze the operation of at least one of the driving assistance function and the automatic driving function.

[0013] The configuration of a simulator 100 according to an embodiment of the present invention (FIG. 1) and detailed processing contents (FIG. 2) will be described below with reference to the two drawings.

[0014] 1 is a block diagram showing an example of the overall configuration of a simulator 100 having a function for analyzing the operation quality of at least one of a driving assistance function and an autonomous driving function. Although the simulator 100 shown in FIG. 1 is constructed as a single computer, it may also operate on a virtual computer constructed on multiple physical computer resources. For example, parallel processing or division of processing by function may be performed.

[0015] The computer that constitutes the simulator 100 includes a processor (CPU), a memory, an auxiliary storage device, and a communication interface. This computer may also include an input / output interface.

[0016] A processor is a computing device that executes programs stored in memory. The processor executes various programs to realize various functional units of a computer. Note that some of the processing performed by the processor by executing the programs may be executed by another computing device (for example, hardware such as an ASIC or FPGA).

[0017] The memory includes ROM, which is a non-volatile storage element, and RAM, which is a volatile storage element. ROM stores unchanging programs (e.g., BIOS), etc. RAM is a high-speed, volatile storage element such as DRAM (Dynamic Random Access Memory), and temporarily stores programs executed by the processor and data used during program execution.

[0018] The auxiliary storage device is, for example, a large-capacity, non-volatile storage device such as a magnetic storage device (HDD) or a flash memory (SSD). The auxiliary storage device also stores data used by the processor when executing a program and the program executed by the processor. In other words, the program is read from the auxiliary storage device, loaded into memory, and executed by the processor to realize each function of the computer (simulator 100).

[0019] The communication interface is a network interface device that controls communication with other devices according to a predetermined protocol.

[0020] The input interface is an interface to which input devices such as a keyboard and a mouse are connected and which receives input from an operator. The output interface is an interface to which output devices such as a display device and a printer (not shown) are connected and which outputs the results of program execution in a format that can be viewed by the operator. Note that a terminal connected via a network may provide the input interface and the output interface.

[0021] The program executed by the processor is provided to the computer from removable media (such as a CD-ROM or flash memory) or via a network, and is stored in a non-volatile auxiliary storage device, which is a non-transitory storage medium. For this reason, the computer should have an interface for reading data from removable media.

[0022] The simulator 100 generates a virtual space based on scenario parameters registered in a reproduction scenario database 103 generated by a reproduction scenario generation unit 102 using test results 101, and can perform a vehicle simulation within that virtual space.

[0023] The test results 101 may be large-scale simulation log data or log data acquired during actual driving.

[0024] Scenario parameters include, for example, spatial parameters such as road (shape, road surface condition), type and position of features (buildings, signs, trees, etc.), weather, time of day, etc., as well as parameters related to traffic participants such as type, position, and speed of vehicles, pedestrians, bicycles, etc.

[0025] The reproduction scenario database 103 is a database in which scenario parameters are registered for each scenario. In the example shown in Fig. 1, a method for generating a reproduction scenario based on test results 101 is described, but a scenario using scenario parameters designed by a user without using the test results 101 may also be registered.

[0026] In the following embodiment, a method for generating the reproduction scenario database 103 using the test results 101 shown in FIG. 1 will be described.

[0027] The reproduction scenario generation unit 102 acquires information on scenario parameters required for generating a reproduction scenario from the contents of the log data of the test results 101, converts the information into a format that allows simulation to be executed, and generates a reproduction scenario database (S201). For example, based on the results of a traffic flow simulation that reproduces traffic phenomena occurring on a road in a virtual road space, the reproduction scenario generation unit 102 creates a scenario that describes scenario parameters in chronological order up to the point of operation of at least one of the vehicle's driving assistance function and autonomous driving function. Similarly, to reproduce the occurrence of a vehicle accident, the reproduction scenario generation unit 102 creates a scenario that describes scenario parameters in chronological order up to the point of the accident.

[0028] For example, when generating a scenario of a scene in which at least one of the driving assistance function and the autonomous driving function is activated, the control activation signal of at least one of the driving assistance function and the autonomous driving function recorded in the test result 101 and the timestamp Ta at that time are obtained. Then, the period of one scenario to be generated is set based on the timestamp Ta. For example, if the start time and end time of one scenario to be generated are T0 and T1, respectively, one scenario is generated so that the following relational expression (1) holds: T0 ≦ Ta ≦ T1 ... (1) T0 and T1 can be freely determined depending on the purpose and application of the simulation. However, as an example of how to determine T0 and T1, they may be set to a value obtained by adding a predetermined time Tb to Ta as shown in expression (2): T0 = Ta - Tb ≦ Ta ≦ T1 = Ta + Tb ... (2) Alternatively, a predetermined time may be used within a range that satisfies expression (1).

[0029] The evaluation target simulation execution unit 104 applies the reproduction scenario in the reproduction scenario database 103 to the virtual space and executes a simulation using a vehicle model equipped with the evaluation target software 105. The results of the simulation execution by the evaluation target software 105 are saved in the log 106 (S202).

[0030] The software 105 to be evaluated is software that is implemented with at least one of a driving assistance function and an autonomous driving function, and includes at least one of a preceding vehicle following function, an automatic braking function for collision avoidance, an automatic steering avoidance function, and a lane departure prevention function.

[0031] Next, the analysis unit 107 has a function of analyzing the operation of at least one of the driving assistance function and the autonomous driving function in the software 105 to be evaluated and storing the results in the analysis result database 114, and includes a comparison simulation execution unit 108, a result comparison unit 111, and an operation operation analysis unit 112.

[0032] The analysis unit 107 determines whether or not to run the analysis unit 107 depending on whether the user selected "with analysis" or "without analysis" when running the simulation (S203). If "with analysis" is selected, the execution of the analysis unit 107 continues, and if "without analysis" is selected, only the evaluation object simulation execution unit 104 is executed and the simulation is completed.

[0033] The comparison simulation execution unit 108 executes a simulation using a vehicle model equipped with result comparison software 109 in a virtual space under the same conditions (same reproduction scenario) as when the evaluation target simulation execution unit 104 executed the simulation (S204). The simulation execution results of the result comparison software 109 are saved in a log 110.

[0034] The result comparison software 109 is software that implements at least one of the driving assistance function and the automatic driving function similar to the evaluation target software 105, and serves as a comparison target for analyzing the performance of the evaluation target software 105.

[0035] The result comparison software 109 is, for example, software before changing the control parameters (adjustment parameters such as control activation timing, control intervention amount, etc.) set in the evaluation target software 105. Alternatively, it may be software in which parameters are set so that the control does not activate in the first place.

[0036] The result comparison unit 111 compares the simulation results using the log 106 that stores the simulation execution results of the evaluation target software 105 and the log 110 that stores the simulation execution results of the result comparison software 109 (S205). For example, when focusing on a collision event, the result of whether or not there was a collision is compared with the information contained in the logs, such as the result of whether or not there was a collision, the speed at the time of the collision, the collision position (fore, to the side, rear, etc. of the vehicle), and the control activation time of at least one of the driving assistance function and the autonomous driving function.

[0037] The operation quality analysis unit 112 analyzes the operation quality of at least one of the driving assistance functions and the autonomous driving functions implemented in the software 105 to be evaluated in accordance with the analysis rules 113 based on the comparison results of each simulation result by the result comparison unit 111 (S207).

[0038] The analysis result database 114 stores the results of the analysis performed in step S207 (S208).

[0039] In step S209, it is determined whether or not there is a setting to execute multiple simulations for result comparison. If other result comparison software 109 is registered in the simulator 100, the processing of the analysis unit 107 (S204 to S208) described above is similarly executed.

[0040] The analysis rule 113 is a rule that defines a judgment on a difference in the comparison result. Some examples of the analysis rule 113 are described below.

[0041] 3 is a diagram showing definitions of analysis rules 113 related to collision scenarios. A collision scenario is a scenario included in the re-enactment scenario database 103, and is a scenario in which the host vehicle and a target collide without the driving assistance function or the autonomous driving function operating, or a scenario in which the host vehicle and a target pass each other on the verge of colliding (but do not actually collide).

[0042] In order to analyze the simulation results for a collision scenario, first, using the comparison results of whether or not a collision occurred, it is determined whether the operation of at least one of the driving assistance functions and automatic driving functions of the software 105 to be evaluated is one of the four types: "normal operation," "abnormal operation," "unnecessary operation," or "determination based on the severity of the collision."

[0043] If the result of the result comparison software 109 is "collision occurred" and the result of the evaluation target software 105 is "no collision," the performance of at least one of the driving assistance function and the automatic driving function of the evaluation target software 105 is improved compared to the performance of the driving assistance function and the automatic driving function of the result comparison software 109, and therefore the analysis result is determined to be "normal operation."

[0044] If the result of the result comparison software 109 is "no collision" and the result of the evaluation target software 105 is "collision," the performance of at least one of the driving assistance functions and automatic driving functions of the evaluation target software 105 is degraded compared to the performance of the driving assistance functions and automatic driving functions of the result comparison software 109, and the analysis result is determined to be "abnormal operation."

[0045] If the result of the result comparison software 109 is "no collision" and the result of the evaluation target software 105 is "no collision," the operation of at least one of the driving assistance function and the automatic driving function of the evaluation target software 105 is unnecessary, so the analysis result is determined to be "unnecessary operation."

[0046] If the result of the result comparison software 109 is "collision occurred" and the result of the software 105 to be evaluated is "collision occurred," it is possible that there is a functional limit in at least one of the driving assistance function and the automatic driving function of the software 105 to be evaluated, and since it is not possible to accurately determine whether or not it is operating properly, the primary analysis result is determined to be "determination based on the degree of collision," and additional analysis is subsequently performed.

[0047] 4 is a diagram showing the definition of the rules for additional analysis when it is determined that "determination based on the degree of collision" is to be performed. An example of additional analysis using the table shown in FIG.

[0048] If the software 105 to be evaluated does not operate collision avoidance control using either the driving assistance function or the automatic driving function, the analysis result will be determined to be "not operating" regardless of whether the avoidance control of the result comparison software 109 is operating or not.

[0049] Focusing on the collision speed of the own vehicle when the own vehicle collides with the target, if the collision speed in the evaluation software 105 is higher than the collision speed in the result comparison software 109, the collision damage is greater, so the collision avoidance control intervention is insufficient and the analysis result is determined to be "performance degradation." On the other hand, if the collision speed in the evaluation software 105 is lower than the collision speed in the result comparison software 109, the collision damage is reduced and the analysis result is determined to be "performance improvement."

[0050] When the relative distance between the vehicle and the target when the collision avoidance control is activated is taken into consideration, if the relative distance when the collision avoidance control is activated by the software 105 to be evaluated is shorter than the relative distance when the collision avoidance control is activated by the software 109 for comparing results, the timing of activation of the collision avoidance control is delayed, and the analysis result is determined to be "performance degradation."

[0051] When the collision position on the own vehicle side when the own vehicle collides with the target (which part of the own vehicle the target collided with) is taken into consideration, for example, if the collision position in the result comparison software 109 is near the hood of the own vehicle, but in the evaluation software 105 the collision occurs in a location close to the occupants of the own vehicle (for example, near the driver's seat), the damage to the occupants will be great, and the analysis result will be determined to be "performance degradation."

[0052] 5 is a diagram showing the definition of the analysis rule 113 related to the lane keeping scenario. The lane keeping scenario is a scenario included in the re-enactment scenario database 103, and is a scenario for testing whether the vehicle can keep within the lane without deviating from the lane by controlling at least one of the driving assistance function and the autonomous driving function.

[0053] In order to analyze the simulation results in the lane keeping scenario, first, using the comparison results of whether or not lane departure occurred, it is determined whether the operation of at least one of the driving assistance functions and autonomous driving functions of the software 105 for evaluation is "normal operation," "abnormal operation," "determination based on the degree of lane departure," or "degree of lane sway."

[0054] If the result of the result comparison software 109 is "Lane departure occurred" and the result of the software to be evaluated 105 is "No lane departure," the operation of at least one of the driving assistance functions and the automatic driving functions of the software to be evaluated 105 has improved performance compared to the operation of at least one of the driving assistance functions and the automatic driving functions of the result comparison software 109, and therefore the analysis result is determined to be "normal operation."

[0055] If the result of the result comparison software 109 is "no lane departure" and the result of the evaluation target software 105 is "lane departure," the performance of at least one of the driving assistance functions and automatic driving functions of the evaluation target software 105 is inferior to the performance of at least one of the driving assistance functions and automatic driving functions of the result comparison software 109, and the analysis result is determined to be "abnormal operation."

[0056] If the result of the result comparison software 109 is "Lane departure occurred" and the result of the software 105 to be evaluated is "Lane departure occurred," it is possible that there is a functional limit in at least one of the driving assistance function and the automatic driving function of the software 105 to be evaluated, and since it is not possible to accurately determine whether or not it is operating properly, the primary analysis result is determined to be "determination based on the degree of lane departure," and additional analysis is then performed.

[0057] 6 is a diagram showing the definition of rules for additional analysis when it is determined that "determination is based on the degree of lane departure." An example of additional analysis using the table shown in FIG.

[0058] Focusing on the amount of lane departure, if the amount of lane departure in the evaluation software 105 is greater than the amount of lane departure in the result comparison software 109, the analysis result is determined to be "performance degradation" because there is insufficient steering correction intervention by at least one of the driving assistance function and the autonomous driving function. On the other hand, if the amount of lane departure in the evaluation software 105 is smaller than the amount of lane departure in the result comparison software 109, the analysis result is determined to be "performance improvement."

[0059] If lane keeping control using either the driving assistance function or the autonomous driving function is not activated in the software 105 to be evaluated, the analysis result will be determined to be "not activated" regardless of whether the avoidance control of the result comparison software 109 is activated or not.

[0060] In FIG. 5, if the result of the result comparison software 109 is "no lane departure" and the result of the evaluation software 105 is "no lane departure," there is a possibility that the evaluation software 105 has not detected lane departure but is driving while swaying within the lane, so the primary analysis result is determined to be "determination based on the degree of swaying within the lane," and additional analysis is subsequently performed.

[0061] 7 is a diagram showing the definition of rules for additional analysis when it is determined that "determination is based on the degree of lane sway." An example of additional analysis using the table shown in FIG.

[0062] First, a method for measuring the degree of lane sway will be described. In the simulation, position information of the lane markings on both sides of the lane in which the vehicle is traveling can be obtained. This makes it possible to calculate the trajectory (e.g., the center of the lane) on which the vehicle should travel. The amount of lateral movement, Diff_trajectory, is calculated, which is the difference between a certain coordinate on the calculated trajectory and the position coordinate of the vehicle (various positions can be used, such as the center, tire position, or leading edge position). The maximum value of Diff_trajectory during the simulation, Diff_trajectory_max, is obtained. By comparing Diff_trajectory_max of the evaluation software 105 with Diff_trajectory_max of the result comparison software 109, the degree of lane sway can be determined, allowing for additional analysis.

[0063] If the maximum value of the lateral movement amount Diff_trajectory_max of the evaluation target software 105 is larger than the maximum value of the lateral movement amount Diff_trajectory_max of the result comparison software 109, the control intervention of the evaluation target software 105 is excessive, and the analysis result is determined to be "performance degradation." On the other hand, if the maximum value of the lateral movement amount Diff_trajectory_max of the evaluation target software 105 is smaller than the maximum value of the lateral movement amount Diff_trajectory_max of the result comparison software 109, the analysis result is determined to be "performance improvement."

[0064] 8 is a diagram showing the definition of the analysis rule 113 related to the merging scenario. The merging scenario is a scenario included in the reproduction scenario database 103, and is a scenario for testing whether the vehicle can smoothly merge onto a main road by controlling at least one of the driving assistance function and the autonomous driving function when merging onto an expressway or a main road.

[0065] In FIG. 8, an analysis rule 113 is defined that aims to achieve smooth merging without sudden deceleration when merging.

[0066] The presence or absence of sudden deceleration, which is information required by the analysis rule 113 in FIG. 8, can be determined by whether the maximum deceleration of the host vehicle during the execution of the simulation exceeds a set threshold value.

[0067] If the result of the result comparison software 109 is "sudden deceleration" and the result of the software to be evaluated 105 is "no sudden deceleration," the performance of at least one of the driving assistance function and the automatic driving function of the software to be evaluated 105 has improved, so the analysis result is determined to be "normal operation."

[0068] If the result of the result comparison software 109 is "no sudden deceleration" and the result of the evaluation software 105 is "sudden deceleration", the performance of at least one of the driving assistance function and the automatic driving function of the evaluation software 105 has deteriorated, so the analysis result is determined to be "abnormal operation".

[0069] If the result of the result comparison software 109 is "no sudden deceleration" and the result of the software 105 to be evaluated is "no sudden deceleration," the operation of at least one of the driving assistance function and the automatic driving function of the software 105 to be evaluated is appropriate, so the analysis result is determined to be "normal operation."

[0070] If the result of the result comparison software 109 is "sudden deceleration occurred" and the result of the software 105 for evaluation is "sudden deceleration occurred," it is possible that there is a functional limit in at least one of the driving assistance function and the automatic driving function of the software 105 for evaluation, and since it is not possible to accurately determine whether the operation is correct or not, the primary analysis result is determined to be "determination based on the degree of sudden deceleration," and additional analysis is subsequently performed.

[0071] 9 is a diagram showing the definition of the rules for additional analysis when it is determined that "judgment is based on the degree of sudden deceleration." An example of additional analysis using the table shown in FIG.

[0072] Focusing on the maximum deceleration, if the maximum deceleration in the evaluation software 105 is greater than the maximum deceleration in the result comparison software 109, smooth merging control is not possible, and the analysis result is determined to be "performance degradation." On the other hand, if the maximum deceleration in the evaluation software 105 is smaller than the maximum deceleration in the result comparison software 109, the analysis result is determined to be "performance improvement."

[0073] The above describes the configuration and processing of a simulation environment equipped with the analysis unit 107, but the specific configuration is not limited to the above-mentioned embodiment, and even if there are design changes, etc., within the scope that does not deviate from the gist of the present invention, they are included in the present invention.

[0074] According to the embodiment of the present invention, the results of a large-scale simulation are automatically analyzed, thereby reducing the number of steps required for manual analysis and improving verification efficiency. Furthermore, a traffic flow simulation that virtually simulates real-world road environments and traffic participants can quickly execute long-term, long-distance driving tests required for actual vehicle testing, and the results can be analyzed by the simulator 100 of the embodiment, making it possible to identify problems in developed driving assistance functions and autonomous driving functions before mass production, thereby reducing recalls.

[0075] Here, as an example of applying the present invention to another configuration, a configuration in which the present invention is applied to the management server 115 is conceivable.

[0076] FIG. 10 is a diagram showing an example of a vehicle control system in which a simulator 100 equipped with the analysis functions described above is installed on a management server 115 to provide a variety of services related to at least one of driving assistance functions and autonomous driving functions.

[0077] The management server 115 has a simulator 100 having the analysis function of the present invention, an information generation unit 116 that uses the results to generate information to be distributed to service recipients (actual vehicles, insurance services, etc.), an information transmission unit 117 that distributes information to service recipients, an information receiving unit 118 that acquires information necessary for providing the service, and a vehicle information database 123 that stores the information acquired by the information receiving unit 118.

[0078] The operation of the simulator 100 has been described above, and therefore will not be described here.

[0079] 10 has an improvement rate calculation unit 119 and a fault ratio calculation unit 120, and generates information to be distributed from the management server 115. In addition to the information described above, the information generation unit 116 may generate various information according to the service provider.

[0080] The improvement rate calculation unit 119 calculates an improvement rate by comparing the results of the evaluation target software 105 analyzed by the simulator 100 with the results of the result comparison software 109. The improvement rate can be calculated as the ratio of the number of scenarios that obtained improved results to the total number of scenarios executed. For example, if there are 1,000 execution scenarios in total and performance improvement results are obtained in 100 of the execution scenarios, the improvement rate is 10%. On the other hand, if there are 1,000 execution scenarios in total and performance degradation results are obtained in 100 of the execution scenarios, the improvement rate is -10%.

[0081] Furthermore, the improvement rate may be calculated by combining the overall results of the evaluated scenarios, or may be calculated for each category of the evaluated scenarios (collision scenario, lane keeping scenario, merging scenario, etc.) Based on this improvement rate information, it is possible to select target vehicles when distributing the evaluation target software 105 to actual vehicles, thereby improving the quality of the software update service for the driving assistance function and the autonomous driving function.

[0082] The fault ratio calculation unit 120 calculates information that contributes to the calculation of the fault ratio, which serves as an index when a car accident occurs. For example, when a car accident occurs with the driver assistance function or the autonomous driving function activated on the own vehicle, an index for determining whether the accident was avoidable or not can be obtained. This is done by setting control parameters in the result comparison software 109 that turn off both the driver assistance function and the autonomous driving function, and then performing a simulation, making it possible to determine whether the accident occurred due to at least one of the driving assistance function and the autonomous driving function being activated.

[0083] For example, if the vehicle speed at the time of collision in the comparison simulation (with both the driving assistance function and the autonomous driving function OFF) is 40 km / h and the vehicle speed at the time of collision in the evaluation simulation (with at least one of the driving assistance function and the autonomous driving function ON) is also 40 km / h, it is considered that the driving assistance function and the autonomous driving function were not the cause of the accident, and the fault ratio of the manufacturer of the driving assistance function and the autonomous driving function will be set low.

[0084] On the other hand, if a collision does not occur in the comparison simulation (where both the driving assistance function and the autonomous driving function are OFF) but a collision does occur in the evaluation simulation (where at least one of the driving assistance function and the autonomous driving function is ON), it is considered that the driving assistance function and the autonomous driving function should not have been activated in the first place, and the manufacturer of the driving assistance function and the autonomous driving function will be given a high percentage of fault.

[0085] The information transmission unit 117 has a software update information acquisition unit 121 and a transmission unit 122, selects a service provision target, and transmits information.

[0086] The software update information acquisition unit 121 selects a target for updating software related to at least one of the driving assistance function and the autonomous driving function. For example, the software update target may be selected using the improvement rate for each scenario calculated by the improvement rate calculation unit 119 and vehicle information stored in the vehicle information database 123 that accumulates information acquired by the information receiving unit 118. The transmission unit 122 transmits software to the vehicle selected as the target for the software update.

[0087] More specifically, the improvement rate calculation unit 119 may calculate an overall improvement rate based on the simulation analysis results of all scenarios, and the transmission unit 122 may transmit software with an improvement rate equal to or greater than a predetermined value to all vehicles. Furthermore, for software in which improvement has been seen in a collision scenario, vehicles traveling in urban areas or areas with heavy traffic congestion may be selected as update targets using vehicle information (travel location, planned travel route, locations with frequent near-misses, etc.) acquired by the information receiving unit 118, and the transmission unit 122 may transmit the software in which improvement has been seen to these selected vehicles.

[0088] Vehicles to be subject to software update may be selected based on the age and degree of deterioration of the vehicle, which are calculated based on the vehicle identification number, sensor data, and the like of the vehicle stored in the vehicle information database 123. If at least one piece of software for a driving assistance function and an autonomous driving function that has been tuned to take into account deterioration over time is evaluated by the simulator 100 of this embodiment and the performance analysis results show no problems, the software can be easily distributed to the aging vehicle.

[0089] Furthermore, information on whether the accident was caused by the driving assistance function and the autonomous driving function, calculated by the fault ratio calculation unit 120, may be transmitted to, for example, an insurance company from the transmission unit 122. Although the information transmission unit 117 in Fig. 10 illustrates the vehicle as an example of the transmission destination, the information may be transmitted to a server of the insurance company, or may be transmitted to the insurance company via vehicle communication.

[0090] The present invention is not limited to the above-described embodiments, and includes various modifications and equivalent configurations within the spirit and scope of the appended claims. For example, the above-described embodiments have been described in detail to clearly explain the present invention, and the present invention is not necessarily limited to configurations including all of the described configurations. Furthermore, part of the configuration of one embodiment may be replaced with the configuration of another embodiment. Furthermore, the configuration of another embodiment may be added to the configuration of one embodiment. Furthermore, part of the configuration of each embodiment may be added, deleted, or replaced with other configurations.

[0091] Furthermore, the aforementioned configurations, functions, processing units, processing means, etc. may be realized in part or in whole in hardware, for example by designing them as integrated circuits, or may be realized in software by having a processor interpret and execute a program that realizes each function.

[0092] Information such as programs, tables, and files that realize each function can be stored in a storage device such as a memory, hard disk, or SSD (Solid State Drive), or in a recording medium such as an IC card, SD card, or DVD.

[0093] In addition, the control lines and information lines shown are those that are considered necessary for explanation, and do not necessarily represent all the control lines and information lines that are necessary for implementation. In reality, it can be assumed that almost all components are interconnected.

Claims

1. A simulator for evaluating at least one of a vehicle's driving assistance function and autonomous driving function, comprising: a reproduction scenario generation unit that reproduces and generates a scenario up to a point where at least one of the vehicle's driving assistance function and autonomous driving function is activated, or up to a point where an accident occurs for the vehicle, based on the results of a traffic flow simulation that reproduces traffic phenomena occurring on roads in a virtual road space; an evaluation target simulation execution unit that executes a simulation using a first control parameter that activates at least one of the vehicle's driving assistance function and autonomous driving function, based on the generated reproduction scenario; a comparison simulation execution unit that executes a simulation using a second control parameter different from the first control parameter, based on the generated reproduction scenario; and an operation proper / fault analysis unit that analyzes the proper operation of at least one of the vehicle's driving assistance function and autonomous driving function, based on the simulation results by the evaluation target simulation execution unit and the simulation results by the comparison simulation execution unit.

2. A simulator as claimed in claim 1, characterized in that the second control parameter is a parameter set for comparison with the results of a simulation using the first control parameter.

3. A simulator as described in claim 2, wherein the second control parameter is a parameter that does not activate either the driving assistance function or the autonomous driving function of the vehicle, and the operation validity analysis unit determines the validity of at least one of the driving assistance function and the autonomous driving function of the vehicle.

4. A simulator as described in claim 2, wherein the second control parameter is a parameter that activates at least one of the driving assistance functions and the autonomous driving functions of the vehicle in a state different from that of the first control parameter, and the operation good / bad analysis unit determines the validity of at least one of the driving assistance functions and the autonomous driving functions of the vehicle that are activated by the first control parameter.

5. A simulator as described in claim 1, characterized in that the operation proper / fault analysis unit determines whether at least one of the driving assistance functions and the autonomous driving functions of the vehicle is operating normally, abnormally, not operating, or unnecessary in accordance with preset analysis rules.

6. A vehicle control system that distributes software update information to a vehicle control device, comprising: a management server; and a vehicle control device that communicates with the management server, wherein the management server has: a simulator; a software update information acquisition unit that selects a vehicle to be updated based on a result of analysis by the simulator and vehicle information, and acquires software update information for at least one of a driving assistance function and an automatic driving function of the selected vehicle; and a transmission unit that transmits the update information to the selected vehicle, wherein the simulator has: a reproduction scenario generation unit that reproduces and generates a scenario up to a point where at least one of the driving assistance function and the automatic driving function of the vehicle is activated or a point where an accident occurs for the vehicle, based on a result of a traffic flow simulation that reproduces a traffic phenomenon occurring on a road in a virtual road space; a simulation execution unit for an evaluation target that executes a simulation using a first control parameter that activates at least one of the driving assistance function and the automatic driving function for the vehicle based on the generated reproduction scenario; and a comparison simulation execution unit that executes a simulation using a second control parameter different from the first control parameter based on the generated reproduction scenario. a vehicle control system comprising: an operation proper / fault analysis unit that analyzes the operation proper / fault of at least one of the vehicle's driving assistance function and autonomous driving function based on the simulation results by the evaluation target simulation execution unit and the simulation results by the comparison simulation execution unit; and wherein the vehicle control device updates software for at least one of the driving assistance function and autonomous driving function of the selected vehicle based on the update information transmitted from the management server.

7. A vehicle control system as described in claim 6, characterized in that the software update information acquisition unit selects vehicles for which software is to be updated based on an improvement rate calculated from the results of simulation analysis by the simulator.

8. A vehicle control system as described in claim 6, wherein the management server has a fault ratio calculation unit that calculates the fault ratio in a traffic accident that occurs when at least one of a driving assistance function and an autonomous driving function is activated based on the results of a simulation analysis by the simulator, and the transmission unit transmits information on the fault ratio of a vehicle involved in the traffic accident.

9. A simulation method in which a simulator evaluates at least one of a driving assistance function and an automatic driving function of a vehicle, the simulator having a calculation device that executes predetermined processing and a storage device accessible by the calculation device, the simulation method comprising: a reproduction scenario generation procedure for reproducing and generating a scenario up to a point where at least one of the driving assistance function and the automatic driving function of the vehicle is activated or a point where an accident occurs for the vehicle, based on the results of a traffic flow simulation that reproduces traffic phenomena occurring on a road in a virtual road space; an evaluation target simulation execution procedure for executing a simulation using a first control parameter that activates at least one of the driving assistance function and the automatic driving function of the vehicle, based on the generated reproduction scenario; a comparison simulation execution procedure for executing a simulation using a second control parameter different from the first control parameter, based on the generated reproduction scenario; and an operation proper / improper analysis procedure for analyzing the proper operation of at least one of the driving assistance function and the automatic driving function of the vehicle, based on the simulation results in the evaluation target simulation execution procedure and the simulation results in the comparison simulation execution procedure.

Citation Information

Patent Citations

  • Traffic flow simulation system

    JP2009169912A

  • Driving support device

    JP2009282702A

  • Information processing device, vehicle control method, and information processing system

    WO2020031652A1