SIMULATOR, VEHICLE CONTROL SYSTEM AND SIMULATION METHOD

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 based on the reproduction scenario using a first control parameter to activate a driver assistance function and / or an automatic driving function for a vehicle; a comparison simulation execution unit for executing a simulation based on the reproduction scenario using a second control parameter; and an operational quality analysis unit for analyzing the operational quality of the driver assistance function and / or the automatic driving function of the vehicle based on the simulation result output by the evaluation object simulation execution unit and a simulation result output by the comparison simulation execution unit.

Description

Technical field

[0001] The present invention relates to a simulator. State of the art

[0002] Scenarios in which a vehicle's driver assistance or automatic driving function, such as in a car, is actually used vary considerably depending on a combination of factors, including the shape of a road, the driving environment (which can range from a purely motorway with relatively few obstacles to urban areas with many obstacles and pedestrians), and the weather. The driver assistance and automatic driving functions must therefore be designed to ensure that vehicles can drive safely in each of these scenarios.

[0003] Therefore, testing techniques that use a simulation capable of reproducing various scenarios, such as those mentioned above, are extremely important in the development of driver assistance and automatic driving functions, so techniques for reproducing and testing traffic accidents and near misses in a virtual space have been developed.

[0004] As a prior art in this field, PTL 1 (JP 2008-261991 A) exists, for example. PTL 1 describes a safety assessment device comprising a simulation device for simulating a vehicle accident by causing a vehicle model, representing a model of a vehicle with a virtual occupant sitting on the vehicle, to drive in a virtual road space; a prediction device for predicting an injury to the occupant based on the conditions under which the vehicle accident simulated by the simulation device occurred; and a display device for displaying the injury predicted by the prediction device. List of literature on patent literature

[0005] PTL 1: JP 2008-261991 A Summary of the invention: Technical problem

[0006] The safety assessment device described in PTL 1 predicts the damage states of an occupant sitting on a vehicle by simulating a vehicle accident, by causing the accident reproduction unit to move a vehicle model with a virtual occupant sitting on the vehicle in a virtual road space, but does not assess the performance of the driver assistance function or the automatic driving function.

[0007] Furthermore, the safety assessment device described in PTL 1 only performs a simulation of an evaluation object function, but not a comparative simulation, which is necessary when analyzing the quality of the simulation execution result of the evaluation object function. Therefore, after completion of the simulation, additional work hours are required to analyze whether the performance of the evaluation object function is superior or inferior.

[0008] The present invention therefore relates in particular to a driver assistance function and an automatic driving function, and an objective of the present invention is to provide a simulator that is able to automatically analyze whether the quality of performance of an evaluation object application is superior or inferior by performing, in addition to simulating the evaluation object, a simulation for a comparison baseline at least once or several times. Solution to the problem

[0009] A representative example of the invention disclosed in the present application is defined as follows. More precisely, a simulator for evaluating at least one driver assistance function and one automatic driving function of a vehicle is provided, wherein the simulator comprises: a reproduction scenario generation unit that reproduces and generates a scenario up to a point where at least one of the driver assistance function and the automatic driving function of the vehicle is activated, or up to a point where an accident with the vehicle occurs, based on a result of a traffic flow simulation that reproduces a traffic phenomenon taking place on a road in a virtual road space;an evaluation object simulation execution unit that performs a simulation using a first control parameter configured to operate at least one of the vehicle's driver assistance function and automatic driving function based on the generated reproduction scenario; a comparison simulation execution unit that performs a simulation using a second control parameter that differs from the first control parameter, based on the generated reproduction scenario; and an operational quality analysis unit that analyzes the operational quality of at least one of the vehicle's driver assistance function and automatic driving function based on a simulation result from the evaluation object simulation execution unit and a simulation result from the comparison simulation execution unit. Advantageous effects of the invention

[0010] According to one aspect of the present invention, the efficiency of software testing can be improved. Problems, configurations, and effects that differ from those described above will further become clear from the following description of the embodiments. Brief description of the drawings [ Fig. 1] Fig. Figure 1 is a block diagram illustrating an overall configuration example of a simulator according to an embodiment of the present invention. [ Fig. 2] Fig. Figure 2 is a flowchart of a process that is carried out by the simulator according to the embodiment of the present invention. [ Fig. 3] Fig. Figure 3 is a diagram that specifies definitions of analysis rules in relation to a collision scenario in the embodiment of the present invention. [ Fig. 4] Fig. Figure 4 is a diagram that specifies definitions of rules for an additional analysis in the embodiment of the present invention when the determination result indicates that it is to be determined “based on the degree of collision”. [ Fig. 5] Fig. Figure 5 is a diagram that specifies definitions of analysis rules in relation to a lane keeping scenario in the embodiment of the present invention. [ Fig. 6] Fig. Figure 6 is a diagram that specifies definitions of rules for an additional analysis in the embodiment of the present invention when the determination result indicates that it is to be determined “based on the degree of track deviation”. [ Fig. 7] Fig. Figure 7 is a diagram that specifies definitions of rules for an additional analysis in the embodiment of the present invention when the determination result indicates that it is to be determined “based on the degree of in-track wobble”. [ Fig. 8] Fig. Figure 8 is a diagram that specifies definitions of analysis rules in relation to a merging scenario in the embodiment of the present invention. [ Fig. 9] Fig. Figure 9 is a diagram that specifies definitions of rules for an additional analysis in the embodiment of the present invention when the determination result indicates that it is to be determined “based on the degree of sudden deceleration”. [ Fig. 10] Fig. Figure 10 is a diagram illustrating an example of a vehicle control system according to the embodiment of the present invention. Description of embodiments

[0011] In one embodiment of the present invention, a simulator 100 is described which is capable of analyzing the quality of the operation of at least one driver assistance function and one automatic driving function by generating a reproduction scenario that represents a reproduction of a scenario of the occurrence of a specific event (e.g., activation of the driver assistance function or the automatic driving function due to an accident or near-accident), by using results of a large-scale simulation, by comparing results of a simulation performed with a first control parameter and a simulation performed with a second control parameter, wherein the first and second parameters represent parameters relating to at least one of the driver assistance function or the automatic driving function, by using the generated reproduction scenario.

[0012] A configuration ( Fig. 1) and detailed processing ( Fig. 2) of the simulator 100 according to the embodiment of the present invention are described below with reference to the two drawings.

[0013] Fig. Figure 1 shows a block diagram illustrating a complete configuration example of Simulator 100, which includes a function for analyzing the quality of operation of at least one of the driver assistance function and the automatic driving function. Although the in Fig. While the simulator 100 shown in Figure 1 is configured as a single computer, it can also run on a virtual machine configured with a variety of physical computer resources. For example, it is also possible to implement parallel processing or distribute processing on a per-function basis.

[0014] A computer implementing Simulator 100 further includes a processor (CPU), memory, an auxiliary storage device, and a communication interface. Additionally, the computer may include an input / output interface.

[0015] The processor is a computing device that executes a program stored in memory. The processor executes various programs to implement the computer's functional units. It's important to note that some of the processing performed by the processor executing the program can also be carried out by another computing device (e.g., hardware such as an ASIC and an FPGA).

[0016] The memory further comprises a ROM, which is a non-volatile memory element, and a RAM, which is a volatile memory element. The ROM stores an immutable program (e.g., BIOS) and the like. The RAM is a high-speed, volatile memory element, such as dynamic random-access memory (DRAM), and temporarily stores a program executed by the processor and data used when the program is executed.

[0017] The auxiliary storage device is, for example, a high-capacity non-volatile storage device, such as a hard disk drive (HDD) or solid-state drive (SSD). The auxiliary storage device also stores data used when the processor executes a program, as well as the program itself, which is executed by the processor. In other words, the program implements the computer's functions (Simulator 100) by being read from the auxiliary storage device, loaded into memory, and executed by the processor.

[0018] Furthermore, the communication interface is a network interface device that controls communication with another device according to a predetermined protocol.

[0019] The input interface is, in particular, an interface to which an input device, such as a keyboard or mouse, is connected and which receives input from an operator. The output interface is, furthermore, an interface to which an output device, such as a display or printer (not shown), is connected and which outputs the result of a program's execution in a format visually recognizable by the operator. It should be noted that the input interface and the output interface can be provided by an end device connected via a network.

[0020] The program executed by the processor is provided to the computer via a removable medium (such as a CD-ROM or flash memory) or a network and stored in a non-volatile auxiliary storage device, which may be a non-volatile storage medium. For this purpose, the computer preferably also has an interface for reading data from a removable medium.

[0021] Furthermore, the simulator 100 is able to generate a virtual space based on scenario parameters registered in a reproduction scenario database 103 generated by a reproduction scenario generation unit 102, using a test result 101, and to execute a vehicle simulation in the virtual space.

[0022] Furthermore, test result 101 may refer to large-scale simulation protocol data or to protocol data recorded during actual driving.

[0023] Examples of scenario parameters include spatial parameters, such as the type or position of a road (the shape, the surface texture of a road) or a feature (such as a building, a sign, a tree), the weather and a time zone, and parameters relating to road users, such as the type, position or speed of a vehicle, a pedestrian or a bicycle.

[0024] Furthermore, the reproduction scenario database 103 is a database in which scenario parameters are registered for each scenario. In the example of Fig. 1 illustrates a procedure for generating a reproduction scenario based on test result 101, but it is also possible to register a scenario using scenario parameters designed by a user without using said test result 101.

[0025] In the embodiment described below, a method for generating the reproduction scenario database 103 using the method described in Fig. 1 illustrated test result 101 explained.

[0026] The Reproduction Scenario Generation Unit 102 generates a database of reproduction scenarios by extracting information about scenario parameters required to generate a reproduction scenario from the descriptions in the log data of the test result 101 and converting the information into a format suitable for simulation (S201). For example, to reproduce the operation of at least one of a vehicle's driver assistance function and automatic driving function, the Reproduction Scenario Generation Unit 102 generates a scenario that describes scenario parameters in chronological order up to the point at which at least one of the vehicle's driver assistance function and automatic driving function is activated, based on the result of a traffic flow simulation, which is a reproduction of traffic phenomena occurring on a road in a virtual road space.Similarly, in order to reproduce an accident involving the vehicle, the reproduction scenario generation unit 102 generates a scenario that describes scenario parameters in chronological order up to the point where the accident occurred.

[0027] For example, to generate a scenario of a scene in which at least one of the driver assistance functions and the automatic driving function is activated, a control operating signal from at least one of the driver assistance functions and the automatic driving function, recorded in test result 101, and a timestamp Ta corresponding to the time of the control operating signal are acquired. Then, using the timestamp Ta as a reference, the period of a scenario to be generated is determined. For example, if the start time of a scenario to be generated is designated T0 and the end time as T1, the scenario is generated in such a way that the following relation expression (1) is satisfied. T0≤Ta≤T1

[0028] T0 and T1 can be determined in any way depending on the intended use of the simulation. As an example of how T0 and T1 can be determined, values ​​obtained by adding a predetermined time Tb using Ta as a reference, as given by formula (2), can also be used. T0=Ta−Tb≤Ta≤T1=Ta+Tb

[0029] Alternatively, it is also possible to use a preset time within a range that satisfies formula (1).

[0030] An evaluation object simulation execution unit 104 subsequently applies a reproduction scenario from the reproduction scenario database 103 to the virtual space and executes a simulation on the vehicle side using a vehicle model with evaluation object software 105. The simulation execution result of the evaluation object software 105 is also stored in the protocol 106 (S202).

[0031] The evaluation object software 105 further represents software that implements at least one of the driver assistance function and the automatic driving function, and is software that includes at least one of the following functions: a function for tracking a vehicle ahead, an automatic braking function for collision avoidance, an automatic steering avoidance function and a lane deviation suppression function.

[0032] An analysis unit 107 also provides a function for analyzing the quality of the operation of at least one of the driver assistance function and the automatic driving function in the evaluation object software 105 and for storing the result in an analysis result database 114 and includes a comparison simulation execution unit 108, a result comparison unit 111 and an operational quality analysis unit 112.

[0033] Analysis Unit 107 determines whether it is necessary to execute it, depending on whether the user selected "with analysis" or "without analysis" before running the simulation (S203). If "with analysis" was selected, Analysis Unit 107 is then executed. If "without analysis" was selected, the simulation is terminated after only the Evaluation Object Simulation Execution Unit 104 has been executed.

[0034] A comparison simulation execution unit 108 then performs a simulation in the virtual space (using the same reproduction scenario) under the same conditions as those performed by the evaluation object simulation execution unit 104 (S204), using a vehicle model equipped with result comparison software 109. The simulation execution result of the result comparison software 109 is also stored in a log 110.

[0035] The result comparison software 109 further represents software that implements at least one of the driver assistance function and the automatic driving function that are similar to those of the evaluation object software 105, and serves as a comparison baseline against which the performance of the evaluation object software 105 is analyzed.

[0036] An example of the result comparison software 109 is the evaluation object software 105, before control parameters (parameters for adjustment, such as the time to activate the control and the extent of the control intervention) that are set for it are changed. The result comparison software 109 can also be a software set with parameters that do not initially activate the control.

[0037] The result comparison unit 111 further compares the simulation execution result using protocol 106, which stores the simulation result of the evaluation object software 105, with protocol 110, which stores the simulation execution result of the result comparison software 109 (S205). For example, if focus is placed on a collision event, the following information from at least one of the driver assistance function and the automatic driving function contained in the protocols is compared; the information includes, for example, the results of the presence or absence of a collision, the speed at the time of the collision, the collided position (front, side, rear, and the like of the ego vehicle), and the time at which the control is activated.

[0038] The operational quality analysis unit 112 analyzes the operational quality of at least one of the driver assistance function and the automatic driving function implemented by the evaluation object software 105, based on results of a simulation result comparison performed by the result comparison unit 111, according to an analysis rule 113 (S207).

[0039] The analysis results database 114 stores the result of the analysis (S208) carried out in step S207.

[0040] Step S209 further determines whether there is a setting to run a large number of simulations to compare the results. If another result comparison software 109 is registered in the simulator 100, the processing described above (S204 to S208) of the analysis unit 107 is also repeated in the same way.

[0041] Analysis rule 113 further defines how the difference in the comparison result is to be determined. Some examples of analysis rule 113 are described below.

[0042] Fig. Figure 3 is a diagram that specifies definitions of the analysis rules 113 with respect to a collision scenario. The collision scenario is either a scenario contained in the reproduction scenario database 103, in which the ego vehicle and a target collide with neither the driver assistance function nor the automatic driving function being operated; or a scenario in which the ego vehicle and the target pass each other on the verge of a collision (without actually colliding).

[0043] To analyze simulation results in the collision scenario, the comparison result regarding the presence or absence of a collision is used to determine which of the following four types the operation of at least one of the driver assistance function and the automatic driving function of the evaluation object software 105 corresponds to, where the four types are "normal operation", "abnormal operation", "unnecessary operation" and "to be determined on the basis of the degree of collision".

[0044] If the result of the result comparison software 109 is “colliding” and the result of the evaluation object software 105 is “not colliding”, because the operation of at least one of the driver assistance function and the automatic driving function of the evaluation object software 105 has shown an improvement compared to the operation of the driver assistance function and the automatic driving function of the result comparison software 109, the analysis result is defined as “normal operation”.

[0045] If the result of the result comparison software 109 is “not in conflict” and the result of the evaluation object software 105 is “in conflict” because the operation of at least one of the driver assistance function and the automatic driving function of the evaluation object software 105 shows a deteriorated performance compared to the operation of the driver assistance function and the automatic driving function of the result comparison software 109, the analysis result is, however, defined as “abnormal operation”.

[0046] If the result of the result comparison software 109 is also “non-collision” and the result of the evaluation object software 105 is also “non-collision” because the operation of at least one of the driver assistance function and the automatic driving function of the evaluation object software 105 is unnecessary, the analysis result is defined as “unnecessary operation”.

[0047] If the result of the result comparison software 109 is furthermore "collided" with the result of the evaluation object software 105, since it is possible that the function of at least one of the driver assistance function and the automatic driving function of the evaluation object software 105 has reached its limit, the quality of operation cannot subsequently be correctly determined. Therefore, a preliminary analysis result is defined here as "to be determined based on the degree of collision" and subsequently an additional analysis is carried out.

[0048] Fig. Figure 4 is a diagram that specifies definitions of rules for additional analysis when the determination result is "to be determined based on the degree of collision". An example of additional analysis using the diagram in Fig. The diagram illustrated in section 4 is described below.

[0049] If a collision avoidance control is not activated by either the driver assistance function or the automatic driving function of the evaluation object software 105, the analysis result is also set as "inactivated", regardless of whether the avoidance control of the result comparison software 109 has been activated or not.

[0050] Focusing on the ego vehicle's collision speed at the moment of collision between the ego vehicle and the target, if the collision speed is higher according to the evaluation object software 105 than the collision speed according to the outcome comparison software 109, the collision avoidance control is determined to be inadequate, and the analysis result is set as "Performance Deterioration" because the damage from the collision increases. Conversely, if the collision speed is lower according to the evaluation object software 105 than that determined by the outcome comparison software 109 because the damage from the collision decreases, the analysis result is set as "Performance Improvement".

[0051] With a focus on the relative distance between the ego vehicle and the target with the operation of the collision avoidance control, if the relative distance with the operation of the collision avoidance control in the evaluation object software 105 is less than the relative distance with the operation of the collision avoidance control in the result comparison software 109, because the time of activation of the collision avoidance control is later, the analysis result is defined as "performance deterioration".

[0052] With a focus on the collided position of the ego vehicle (which part of the ego vehicle the target collided with) at the time the ego vehicle and the target collided, if the collided position is near the hood of the ego vehicle according to the result comparison software 109, but the collided position is near the occupant of the ego vehicle according to the evaluation object software 105 (e.g., near the driver's seat), because, for example, the damage received by the occupant is greater, the analysis result is also defined as "performance degradation".

[0053] Fig. Figure 5 shows a diagram that provides definitions of the analysis rules 113 in relation to a lane keeping scenario. The lane keeping scenario is a scenario included in the reproduction scenario database 103 and is a scenario for testing whether the ego vehicle continues to drive within the lane without deviating from it, under the control of at least one of the driver assistance functions and the automatic driving function.

[0054] To analyze the simulation results in the lane keeping scenario, the comparison results obtained by comparing whether a deviation from the lane exists are used to determine which of the four types the operation of at least one of the driver assistance function and the automatic driving function of the evaluation object software 105 corresponds to, where the four types are "normal operation", "abnormal operation", "to be determined on the basis of the degree of lane deviation" or "to be determined on the basis of the degree of in-lane wobbling".

[0055] If the result of the result comparison software 109 has "deviated from the track" and the result of the evaluation object software 105 has "not deviated from the track", because the operation of at least one of the driver assistance function and the automatic driving function of the evaluation object software 105 has shown an improvement compared to the operation of at least one of the driver assistance function and the automatic driving function of the result comparison software 109, the analysis result is defined as "normal operation".

[0056] If the result of the result comparison software 109 is "not off track" and the result of the evaluation object software 105 is "off track" because the operation of at least one of the driver assistance function and the automatic driving function of the evaluation object software 105 shows a deteriorated performance compared to the operation of at least one of the driver assistance function and the automatic driving function of the result comparison software 109, the analysis result is, however, defined as "abnormal operation".

[0057] If the result of the result comparison software 109 subsequently "deviates from the target" and the result of the evaluation object software 105 also "deviates from the target," it is possible that the function of at least one of the driver assistance function and the automatic driving function of the evaluation object software 105 has reached its limit and the quality of operation cannot be correctly determined subsequently. Therefore, a preliminary analysis result is defined here as "to be determined based on the degree of deviation from the target" and a further analysis is then carried out.

[0058] Fig. Figure 6 is a diagram that provides definitions of rules for additional analysis when the determination result is to be determined "based on the degree of trace deviation". An example of additional analysis using the diagram in Fig. The diagram illustrated in section 6 is described below.

[0059] Focusing on the extent of lane deviation, if the extent of lane deviation measured with the evaluation object software 105 is greater than that measured with the result comparison software 109, because the steering correction intervention of at least one of the driver assistance functions and the automatic driving function is insufficient, the analysis result is defined as "performance deterioration". Conversely, if the extent of lane deviation measured with the evaluation object software 105 is less than that measured with the result comparison software 109, the analysis result is defined as "performance improvement".

[0060] If the lane keeping control of neither the driver assistance function nor the automatic driving function of the evaluation object software 105 has been activated, the analysis result will still be set as "inactivated", regardless of whether the avoidance control of the result comparison software 109 has been activated or not.

[0061] If in Fig. 5. If the result of the result comparison software 109 is also "not deviated from the track" and the result of the evaluation object software 105 is likewise "not deviated from the track", it is possible that the vehicle is wobbling with the evaluation object software 105. Although there is no deviation from the track, the preliminary analysis result in this case is defined as "to be determined based on the degree of in-track wobble" and the additional analysis is then carried out.

[0062] Fig. Figure 7 is a diagram that provides definitions of rules for additional analysis when the determination result is to be determined “based on the degree of in-track wobble”. An example of the additional analysis using the diagram in Figure 7 is shown. Fig. The diagram illustrated in section 7 is described below.

[0063] First, a method for measuring the degree of in-lane wobble is described. During the simulation, positional information from lane markings on both sides of the lane in which the ego-vehicle is traveling can be determined. This allows the trajectory (e.g., the center of the lane) along which the ego-vehicle should travel to be calculated. The magnitude of the lateral movement, Diff_trajectory, is calculated; that is, the difference in the coordinates of the ego-vehicle's position (various positions, such as the center, the positions of the tires, and the position of the head, can be used as the position) with respect to a specific coordinate on the calculated trajectory. Subsequently, the maximum Diff_trajectory_max is determined during the simulation.By comparing Diff_trajektorie_max, which is obtained by the evaluation object software 105, with Diff_trajektorie_max, which is obtained by the result comparison software 109, the degree of wobble in the track can be detected and the additional analysis can be carried out.

[0064] If the maximum value Diff_trajektorie_max for the magnitude of lateral movement with evaluation object software 105 is greater than the maximum value Diff_trajektorie_max for the magnitude of lateral movement with result comparison software 109, the control intervention of evaluation object software 105 is excessive, so the analysis result is defined as "performance deterioration". Conversely, if the maximum value Diff_trajektorie_max for the magnitude of lateral movement with evaluation object software 105 is less than the maximum value Diff_trajektorie_max for the magnitude of lateral movement with result comparison software 109, the analysis result is defined as "performance improvement".

[0065] Fig. Figure 8 shows a diagram that specifies definitions of the analysis rules 113 in relation to a merge scenario. The merge scenario is a scenario contained in the reproduction scenario database 103 and is a scenario for testing whether the ego vehicle can smoothly merge into the main line by controlling at least one of the driver assistance function and the automatic driving function at the time of merging into traffic on a highway or motorway.

[0066] Fig. Section 8 defines analysis rules 113 to enable smooth threading without a sudden slowdown at the time of threading.

[0067] The presence of a sudden slowdown, which is information relevant to analysis rule 113 in Fig. The number of necessary parameters can be determined based on whether the maximum deceleration of the Ego vehicle during the simulation exceeds a defined threshold.

[0068] If the result of the result comparison software 109 indicates "with sudden deceleration" and the result of the evaluation object software 105 indicates "without sudden deceleration" because the performance of at least one of the driver assistance function and the automatic driving function of the evaluation object software 105 has improved, the analysis result is set as "normal operation".

[0069] If, however, the result of the result comparison software 109 indicates "without sudden slowdown" and the result of the evaluation object software 105 indicates "with sudden slowdown" because the performance of at least one of the driver assistance function and the automatic driving function of the evaluation object software 105 has deteriorated, the analysis result is defined as "abnormal operation".

[0070] If the result of the result comparison software 109 subsequently indicates "without sudden slowdown" and the result of the evaluation object software 105 also indicates "without sudden slowdown" because the performance of at least one of the driver assistance function and the automatic driving function of the evaluation object software 105 is adequate, the analysis result is also defined as "normal operation".

[0071] Furthermore, if the result of the result comparison software 109 is "with sudden deceleration" and the result of the evaluation object software 105 is also "with sudden deceleration," since it is possible that the function of at least one of the driver assistance function and the automatic driving function of the evaluation object software 105 has reached its limit, the quality of operation cannot be correctly determined. Therefore, in this case, a preliminary analysis result is set as "to be determined based on the degree of sudden deceleration," and subsequently, an additional analysis is performed.

[0072] Fig. Figure 9 is a diagram that specifies definitions of rules for the additional analysis in the embodiment of the present invention when the determination result is to be determined “based on the degree of sudden deceleration”. An example of the additional analysis using the diagram in Figure 9 is shown in Figure 9. Fig. The diagram illustrated in section 9 is described below.

[0073] Focusing on maximum deceleration, if the maximum deceleration in the evaluation object software 105 is greater than the maximum deceleration in the result comparison software 109, the smooth threading control has failed, and therefore the analysis result is set as "performance deterioration." Conversely, if the maximum deceleration with the evaluation object software 105 is less than the maximum deceleration with the result comparison software 109, the analysis result is set as "performance improvement."

[0074] Although the configuration of the simulation environment, which includes the analysis unit 107 and the processing, has been described above, the specific configuration is not limited to the embodiment described above. Even if the design is modified or the like, without departing from the core of the present invention, such a modification or the like still falls within the scope of the present invention.

[0075] Since the results of a large-scale simulation are automatically analyzed, it is possible, according to the embodiment of the present invention, to reduce the hours required for manual analysis and improve testing efficiency. Furthermore, using a traffic flow simulation that virtualizes and simulates the real road environment and road users, driving tests can be performed quickly, which would require longer driving times and distances when using actual vehicles. By performing the analysis of the results using the simulator 100 according to the embodiment, it is also possible to clarify problems in the developed driver assistance function and the automatic driving function before they are introduced into mass production, thus also reducing recalls.

[0076] As another example in which the present invention can be applied to a different configuration, a configuration is also possible in which it is applied to an administration server 115.

[0077] Fig. Figure 10 shows a diagram illustrating an example of a vehicle control system for providing various services relating to at least one of the driver assistance function and the automatic driving function by installing the simulator 100 with the analysis function described above on the management server 115.

[0078] The management server 115 comprises: the simulator 100 with the analysis function according to the present invention; an information generation unit 116, which generates information to be distributed to recipients of the services (such as actual vehicles, insurance services) using a result of the analysis; an information transmission unit 117, which distributes information to the recipients of the services; an information reception unit 118, which acquires information required for the provision of the services; and a vehicle information database 123, which collects information acquired by the information reception unit 118.

[0079] Since the operation of the simulator 100 corresponds to what is described above, its description will be omitted below.

[0080] The in Fig. The illustrated information generation unit 116 further comprises an improvement ratio calculation unit 119 and a comparative negligence calculation unit 120 and generates information to be distributed by the management server 115. In addition to the information described above, the information generation unit 116 can also generate various types of information depending on the recipients of the services.

[0081] The improvement ratio calculation unit 119 calculates an improvement ratio by comparing the result from the evaluation object software 105, where the result is analyzed by the simulator 100, with the result from the result comparison software 109. The improvement ratio can be calculated, in particular, as the ratio of the number of scenarios in which improvement results are achieved to the total number of simulated scenarios. For example, if the total number of simulated scenarios is 1000 and results of a performance improvement are obtained in 100 of such simulated scenarios, the improvement ratio is 10%. Conversely, if the total number of simulated scenarios is 1000 and results of a performance deterioration are obtained in 100 of such simulated scenarios, the improvement ratio is -10%.

[0082] The improvement ratio can be calculated by integrating the results for all evaluated scenarios or for each category of evaluated scenarios (e.g., collision scenarios, lane keeping scenarios, merging scenarios). Based on this improvement ratio information, it is then possible to select the vehicles to which the evaluation object software 105 should be distributed at the time of distribution to the actual vehicles, and to improve the quality of the software update services for the driver assistance function and the automatic driving function.

[0083] The comparative negligence calculation unit 120 further calculates information that contributes to the calculation of a comparative negligence index in the event of a car accident. For example, if the Ego vehicle is involved in a car accident while the driver assistance function or the automatic driving function is active, it is possible to obtain an index to determine whether the accident was avoidable. By running a simulation using the result comparison software 109 with the control parameter to deactivate both the driver assistance function and the automatic driving function, it is accordingly possible to determine whether the accident occurred due to the operation of at least one of the driver assistance function and the automatic driving function.

[0084] For example, if a comparative simulation indicates an ego-vehicle speed (with both the driver assistance function and the automatic driving function set to OFF) at the time of the collision as 40 km / h, and a simulation for the evaluation object also indicates an ego-vehicle speed (with at least one of the driver assistance function and the automatic driving function set to ON) at the time of the collision as 40 km / h, it is assumed that the driver assistance function and the automatic driving function are not the causes of the accident, and the comparative negligence of the manufacturer of the driver assistance function and the automatic driving function is set to be low.

[0085] If, however, no collision occurs in the comparison simulation (with both the driver assistance function and the automatic driving function OFF), and a collision occurs in the simulation for the evaluation object (with at least one of the driver assistance function and the automatic driving function ON), it is assumed that the driver assistance function and the automatic driving function should not have been activated in the first place, and the comparative negligence of the manufacturer of the driver assistance function and the automatic driving function is assumed to be high.

[0086] The information transmission unit 117 also includes a software update information acquisition unit 121 and a transmission unit 122, and selects the recipient of the services and transmits information to it.

[0087] The software update information acquisition unit 121 selects a target for a software update relating to at least one of the driver assistance functions and the automatic driving function. For example, a software update recipient can be selected using the improvement ratio corresponding to each scenario and the vehicle information, where the improvement ratio is calculated by the improvement ratio calculation unit 119 and the vehicle information is stored in the vehicle information database 123, which collects the information acquired by the information receiving unit 118. The transmission unit 122 then transmits the software to the vehicles selected as software update recipients.

[0088] In particular, the improvement ratio calculation unit 119 can calculate a comprehensive improvement ratio based on the simulation analysis results across all scenarios, and the transmission unit 122 can transfer software exhibiting an improvement ratio equal to or greater than a predetermined value to all vehicles. Furthermore, for the software that has resulted in improvements in the collision scenarios, vehicles operating in urban areas or areas with frequent traffic jams can be selected as software update targets, using vehicle information (such as current positions, planned routes, and locations where near misses frequently occur) acquired by the information receiving unit 118. The transmission unit 122 can then transfer the software that has demonstrated improvements to the selected vehicles.

[0089] The vehicles to be provided with software updates can also be selected based on the number of years or the degree of deterioration of the vehicles, calculated based on vehicle body numbers, vehicle sensor data, or the like, stored in the vehicle information database 123. By performing evaluations of at least one software program for the driver assistance function and the automatic driving function, which have been adjusted according to the deterioration over time, the process of distributing the software to the deteriorated vehicles can be easily carried out using the simulator 100 according to the embodiment, and if it has been determined that there is no problem in the result of the performance quality analysis.

[0090] Furthermore, the transmission unit 122 can also transmit information about whether the driver assistance function and the automatic driving function caused an accident, or the information calculated by the comparative negligence calculation unit 120, for example to an insurance company. In the example in Fig. 10 are the recipients of the information from the information transmission unit 117, illustrated as an example as vehicles, whereby the information transmission unit 117 can also transmit information to a server of an insurance company or to the insurance company via vehicle communication.

[0091] Furthermore, it should be noted that the present invention is not limited to the embodiments described above and includes various modifications and equivalent configurations as defined in the appended claims. For example, the embodiments have been described in such detail only to facilitate understanding of the present invention, but the present invention is not necessarily limited to configurations that include all the described elements. Moreover, a part of the configuration according to one embodiment can be replaced by the configuration of another embodiment. The configuration according to one embodiment can also include the configuration according to another embodiment. In addition, a part of the configuration according to each of the embodiments can also be added to, deleted from, or replaced by another configuration.

[0092] Furthermore, some or all of the configurations, functions, processing units, processing means and the like described above may be implemented as hardware, for example by designing an integrated circuit, or as software by causing a processor to parse and execute a program to implement the functions.

[0093] Information, such as a program, a table, and a file for implementing these functions, may also be stored in a storage device, such as memory, a hard disk, and a solid-state drive (SSD), or a recording medium, such as an IC card, an SD card, and a DVD.

[0094] Furthermore, only the control and information lines deemed necessary in the description are shown; not all control and information lines required for implementation need to be displayed. In practice, it can also be assumed that almost all of these configurations are interconnected. QUOTES INCLUDED IN THE DESCRIPTION

[0000] This list of documents cited by the applicant was automatically generated and is included solely for the reader's convenience. The list is not part of the German patent or utility model application. The DPMA accepts no liability for any errors or omissions. Cited patent literature

[0000] JP 2008-261991 A [0004, 0005]

Claims

[1] A simulator for evaluating at least one driver assistance function and / or one automatic driving function of a vehicle, wherein the simulator comprises: a reproduction scenario generation unit that reproduces and generates a scenario up to a point where at least one of the vehicle's driver assistance function and automatic driving function is activated, or up to a point where an accident with the vehicle occurs, based on the result of a traffic flow simulation that reproduces a traffic phenomenon taking place on a road in a virtual road space; an evaluation object simulation execution unit that performs a simulation using a first control parameter configured to operate at least one of the vehicle's driver assistance function and automatic driving function based on the generated reproduction scenario; a comparison simulation execution unit that executes a simulation using a second control parameter that differs from the first control parameter, based on the generated reproduction scenario; and An operational quality analysis unit that analyzes the operational quality of at least one of the vehicle's driver assistance functions and automatic driving functions based on a simulation result from the evaluation object simulation execution unit and a simulation result from the comparison simulation execution unit. [2] The simulator according to claim 1, wherein the second control parameter corresponds to a parameter set for comparing simulation results using the first control parameter. [3] The simulator according to claim 2, wherein the second control parameter corresponds to a parameter that does not activate either the driver assistance function or the automatic driving function of the vehicle, and The operational quality analysis unit determines the effectiveness of at least one driver assistance function and / or one automatic driving function of the vehicle. [4] The simulator according to claim 2, wherein the second control parameter corresponds to a parameter to operate at least one of the driver assistance function and / or the automatic driving function of the vehicle in a state that differs from the first control parameter, and the operational quality analysis unit determines an effectiveness of at least one of the driver assistance function and / or the automatic driving function of the vehicle, wherein at least one of them is operated by the first control parameter. [5] The simulator according to claim 1, wherein the operational quality analysis unit determines normal operation, abnormal operation, non-operation and unnecessary operation of at least one of the driver assistance function and / or the automatic driving function of the vehicle according to a preset analysis rule. [6] A vehicle control system that distributes update information from software to a vehicle control device, the vehicle control system comprising: an administration server; and a vehicle control unit that communicates with the management server, whereby The administration server includes: a simulator; a software update information acquisition unit that selects a vehicle whose software is to be updated, based on the result of an analysis performed by the simulator and vehicle information, and that acquires update information from the software of at least one driver assistance function and / or an automatic driving function of the vehicle so selected; and a transmission unit that transmits the update information to the selected vehicle, and The simulator includes: a reproduction scenario generation unit that reproduces and generates a scenario up to a point where at least one of the vehicle's driver assistance function and automatic driving function is activated, or up to a point where an accident with the vehicle occurs, based on the result of a traffic flow simulation that reproduces a traffic phenomenon taking place on a road in a virtual road space; an evaluation object simulation execution unit that performs a simulation using a first control parameter configured to operate at least one of the vehicle's driver assistance function and automatic driving function based on the generated reproduction scenario; a comparison simulation execution unit that executes a simulation using a second control parameter that differs from the first control parameter, based on the generated reproduction scenario; and an operational quality analysis unit that analyzes the operational quality of at least one of the vehicle's driver assistance functions and automatic driving functions based on a simulation result from the evaluation object simulation execution unit and a simulation result from the comparison simulation execution unit, and The vehicle control unit software of at least one of the driver assistance function and the automatic driving function of the selected vehicle is updated based on the update information received from the management server. [7] The vehicle control system according to claim 6, wherein the software update information acquisition unit selects a vehicle in which the software is to be updated, based on an improvement ratio calculated from an analysis result of the simulation by the simulator. [8] The vehicle control system according to claim 6, wherein The management server includes a comparative negligence calculation unit that calculates a comparative negligence in a traffic accident that occurred during the operation of at least one of the driver assistance function and the automatic driving function, based on an analysis result of a simulation by the simulator, and The transmission unit transmits information about the comparative negligence of the vehicle involved in the traffic accident. [9] A simulation method that causes a simulator to evaluate at least one driver assistance function and / or an automatic driving function of a vehicle, wherein the simulator comprises a processor device that performs predetermined processing and a storage device that can be accessed by the processor device, wherein the simulation method comprises: a reproduction scenario generation step to reproduce and generate a scenario up to a point where at least one of the vehicle's driver assistance function and automatic driving function is activated, or up to a point where an accident with the vehicle occurs, based on a result of a traffic flow simulation that reproduces a traffic phenomenon taking place on a road in a virtual road space; an evaluation object simulation execution step to execute a simulation using a first control parameter configured to operate at least one of the vehicle's driver assistance function and automatic driving function based on the generated reproduction scenario; a comparison simulation execution step to execute a simulation using a second control parameter that differs from the first control parameter, based on the generated reproduction scenario; and an operational quality analysis step to analyze the quality of an operation of at least one of the driver assistance function and the automatic driving function of the vehicle based on a simulation result in the evaluation object simulation execution step and a simulation result in the comparison simulation execution step.

Images