Function positioning method, device, storage medium and computer equipment

Abstract

The present application provides a function localization method, apparatus, storage medium, and computer device, the method comprising: determining target difference scenarios for which function localization is required, and the original difference performance between a baseline driving algorithm and a driving algorithm to be compared in each target difference scenario; obtaining pre-set functions to be located; for each function to be located, turning off only the function to be located of the driving algorithm to be compared, and performing simulation tests on the driving algorithm to be compared after the function is turned off using each target difference scenario, so as to obtain autonomous driving data of the function to be located in each target difference scenario; for each target difference scenario, determining the test difference performance corresponding to each function to be located based on the autonomous driving data of each function to be located in the target difference scenario, and determining a target function from each function to be located based on each test difference performance and the original difference performance corresponding to the target difference scenario.

Description

Technical Field

[0001] The present application relates to the field of algorithm testing, and in particular to a function positioning method, apparatus, storage medium, and computer equipment. Background Art

[0002] With the continuous iteration of autonomous driving algorithms, autonomous driving algorithms have more and more functions. In order to evaluate the advantages and disadvantages of the old and new versions of autonomous driving algorithms, the new and old versions of the autonomous driving algorithms can be simulated and tested using complete road section data. Based on the differences in trajectory planning between the different versions of the algorithm in the simulation test, several difference scenarios are extracted from the complete road section data. After obtaining each difference scenario, the tester manually determines which function in the new version of the autonomous driving algorithm causes each difference scenario, that is, the function of each difference scenario is located separately, so as to facilitate subsequent algorithm debugging, evaluation and iteration. Because the existing technology relies entirely on manual function location, it has the problem of low efficiency. Summary of the Invention

[0003] The purpose of this application is to solve at least one of the above technical deficiencies, especially the technical defect of low testing efficiency in the prior art.

[0004] In a first aspect, an embodiment of the present application provides a function positioning method, the method comprising:

[0005] Determining each target difference scenario for which functional positioning is required, and the original difference performance between the baseline driving algorithm and the driving algorithm to be compared in each target difference scenario;

[0006] Get the preset functions to be located;

[0007] For each of the functions to be located, only the function to be located of the driving algorithm to be compared is disabled, and simulation tests are performed on the driving algorithm to be compared with the disabled function using each of the target difference scenarios, so as to obtain autonomous driving data of the function to be located in each of the target difference scenarios;

[0008] For each of the target difference scenarios, based on the autonomous driving data of each of the functions to be positioned in the target difference scenario, the test difference performance corresponding to each of the functions to be positioned is determined respectively, and based on each of the test difference performances and the original difference performance corresponding to the target difference scenario, the target function is determined from each of the functions to be positioned, and the target function is the function that causes the original difference performance of the target difference scenario.

[0009] In one embodiment, the step of determining each target difference scenario requiring functional positioning, and the original difference performance between the baseline driving algorithm and the driving algorithm to be compared in each target difference scenario, includes:

[0010] Obtaining each of the original difference scenarios, where the original difference scenarios are scenarios in which the difference in planned trajectories between the baseline driving algorithm and the driving algorithm to be compared satisfies a preset rule;

[0011] Obtain various pre-set difference evaluation indicators;

[0012] For each of the difference evaluation indicators, an initial difference performance for each of the original difference scenarios under the difference evaluation indicator, which is used to reflect the degree of difference between the baseline driving algorithm and the driving algorithm to be compared, is determined. Based on the initial difference performances corresponding to the difference evaluation indicator, original difference scenarios with a large degree of difference are screened out from the original difference scenarios as target difference scenarios, and the initial difference performance corresponding to each target difference scenario is used as the original difference performance corresponding to the target difference scenario.

[0013] In one embodiment, each of the difference evaluation indicators includes an error reporting difference indicator, and each of the target difference scenarios includes a first target difference scenario;

[0014] The step of determining, for each of the original difference scenarios, an initial difference performance reflecting the degree of difference between the baseline driving algorithm and the driving algorithm to be compared under the difference evaluation index, and screening, based on each of the initial difference performances corresponding to the difference evaluation index, an original difference scenario with a large degree of difference from each of the original difference scenarios as a target difference scenario, includes:

[0015] respectively obtaining first error information of the benchmark driving algorithm in each of the original difference scenarios;

[0016] Respectively obtaining second error information of the driving algorithm to be compared in each of the original difference scenarios;

[0017] determining, based on the first error reporting information and the second error reporting information, an initial difference performance under the error reporting difference indicator for each of the original difference scenarios, the initial difference performance being used to reflect whether there is an error reporting difference between the baseline driving algorithm and the driving algorithm to be compared under the corresponding original difference scenario;

[0018] According to the initial difference performance of each of the original difference scenes under the error reporting difference indicator, each of the original difference scenes with error reporting differences is selected as the first target difference scene.

[0019] In one embodiment, the step of determining the target function from each of the functions to be located based on each of the test difference performances and the original difference performance corresponding to the target difference scenario includes:

[0020] If the target difference scenario is the first target difference scenario, the test difference performance that is identical to the original difference performance corresponding to the first target difference scenario in each of the test difference performances is taken as the target test difference performance, and the function to be located corresponding to each of the target test difference performances is determined as the target function.

[0021] In one embodiment, each of the difference evaluation indicators includes a simulation score difference indicator, and each of the target difference scenarios includes a second target difference scenario;

[0022] The step of determining, for each of the original difference scenarios, an initial difference performance reflecting the degree of difference between the baseline driving algorithm and the driving algorithm to be compared under the difference evaluation index, and screening, based on each of the initial difference performances corresponding to the difference evaluation index, an original difference scenario with a large degree of difference from each of the original difference scenarios as a target difference scenario, includes:

[0023] respectively obtaining a first simulation score of the benchmark driving algorithm in each of the original difference scenarios;

[0024] respectively obtaining a second simulation score of the driving algorithm to be compared under each of the original difference scenarios;

[0025] For each of the original difference scenarios, calculating the difference between the first simulation score corresponding to the original difference scenario and the second simulation score corresponding to the original difference scenario, and using the difference as the initial difference performance corresponding to the original difference scenario under the simulation score difference indicator;

[0026] The initial difference performances of each of the original difference scenarios under the simulation score difference index are sorted in order from large to small, and the original difference scenarios corresponding to the first N initial difference performances are selected as the second target difference scenarios, where N is a preset positive integer.

[0027] In one embodiment, each of the difference evaluation indicators includes a driving behavior difference indicator, and each of the target difference scenarios includes a third target difference scenario;

[0028] The step of determining, for each of the original difference scenarios, an initial difference performance reflecting the degree of difference between the baseline driving algorithm and the driving algorithm to be compared under the difference evaluation index, and screening, based on each of the initial difference performances corresponding to the difference evaluation index, an original difference scenario with a large degree of difference from each of the original difference scenarios as a target difference scenario, includes:

[0029] respectively obtaining a first vehicle state of the benchmark driving algorithm in each of the original difference scenarios;

[0030] respectively obtaining a second vehicle state of the driving algorithm to be compared under each of the original difference scenarios;

[0031] For each of the original difference scenarios, generating a difference score reflecting the degree of difference in driving behavior between the baseline driving algorithm and the driving algorithm to be compared based on a first vehicle state corresponding to the original difference scenario and a second vehicle state corresponding to the original difference scenario, and using the difference score as an initial difference performance corresponding to the original difference scenario under the driving behavior difference index;

[0032] The initial difference performances of each of the original difference scenarios under the driving behavior difference index are sorted in order from large to small, and the original difference scenarios corresponding to the first N initial difference performances are selected as the third target difference scenarios, where N is a preset positive integer.

[0033] In one embodiment, the step of determining the target function from each of the functions to be located based on each of the test difference performances and the original difference performance corresponding to the target difference scenario includes:

[0034] If the target difference scenario is the second target difference scenario, determining a screening interval according to the original difference performance corresponding to the second target difference scenario, the screening interval including the original difference performance;

[0035] If the target difference scenario is the third target difference scenario, determining a screening interval according to the original difference performance corresponding to the third target difference scenario, the screening interval including the original difference performance;

[0036] Target test difference performances falling within the screening interval are screened out from each of the test difference performances, and the target function is determined according to the to-be-located functions corresponding to each of the target test difference performances.

[0037] In a second aspect, an embodiment of the present application provides a function positioning device, the device comprising:

[0038] A scenario determination module is used to determine each target difference scenario for which function positioning is required, and the original difference performance between the baseline driving algorithm and the driving algorithm to be compared in each target difference scenario;

[0039] Function acquisition module, used to obtain each pre-set function to be located;

[0040] a simulation module configured to, for each of the functions to be located, disable only the function to be located of the driving algorithm to be compared, and perform simulation tests on the driving algorithm to be compared after the function is disabled using each of the target difference scenarios, so as to obtain autonomous driving data of the function to be located in each of the target difference scenarios;

[0041] A positioning module is used to determine, for each target difference scenario, the test difference performance corresponding to each of the functions to be positioned based on the autonomous driving data of each of the functions to be positioned in the target difference scenario, and to determine a target function from each of the functions to be positioned based on each of the test difference performances and the original difference performance corresponding to the target difference scenario, where the target function is the function that causes the original difference performance of the target difference scenario.

[0042] In a third aspect, an embodiment of the present application provides a storage medium storing computer-readable instructions. When the computer-readable instructions are executed by one or more processors, the one or more processors execute the steps of the function positioning method described in any of the above embodiments.

[0043] In a fourth aspect, an embodiment of the present application provides a computer device, comprising: one or more processors, and a memory;

[0044] The memory stores computer-readable instructions, and when the computer-readable instructions are executed by the one or more processors, the steps of the function positioning method described in any of the above embodiments are executed.

[0045] In the function positioning, apparatus, storage medium, and computer device of the present application, the computer device can determine the target difference scenario for function positioning and the original difference performance of two different versions of autonomous driving algorithms in each target difference scenario, and obtain each pre-set function to be positioned. For each function to be positioned, the computer device can turn off the function to be positioned of the driving algorithm to be compared, and keep the other functions turned on, and use each target difference scenario to simulate and test the driving algorithm to be compared after the function is turned off, so as to obtain autonomous driving data of the function to be positioned in each target difference scenario. For each target difference scenario, the computer device can determine the test difference performance of the baseline driving algorithm and the driving algorithm to be compared when one of the functions to be positioned is turned off in each target difference scenario based on the autonomous driving data of each function to be positioned in the target difference scenario, and based on the original difference performance corresponding to the target difference scenario and each test difference performance, locate the function that causes the original difference performance from each function to be positioned as the target function. In this way, computer equipment can automatically locate the differences between two different versions of autonomous driving algorithms to one or more specific functional features, allowing engineers to perform manual verification based on the results of the location. This can greatly reduce the number of functions that require manual screening and location, thereby reducing the human resources consumed and improving testing efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0047] Figure 1 1 is a flow chart of a function positioning method in one embodiment;

[0048] Figure 2 A flowchart illustrating steps for determining target difference scenarios and original difference representations in one embodiment;

[0049] Figure 3 FIG1 is a flow chart of selecting a target difference scene from various original difference scenes in one embodiment;

[0050] Figure 4 FIG2 is a second flow chart of screening target difference scenes from various original difference scenes in one embodiment;

[0051] Figure 5 FIG3 is a flowchart of screening target difference scenes from various original difference scenes in one embodiment;

[0052] Figure 6 is a structural block diagram of a function positioning device in one embodiment;

[0053] Figure 7 FIG. 1 is a schematic diagram of the structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0054] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0055] In one embodiment, the present application provides a method for function positioning. The following embodiments are described using the method applied to a computer device as an example. It can be understood that the computing device refers to a device with data processing capabilities, which can be, but is not limited to, a personal laptop, a desktop laptop, a single server, or a server cluster. Figure 1 As shown, the method of the present application may include the following steps:

[0056] S102: Determine each target difference scenario for which function positioning is required, and determine the original difference performance between the baseline driving algorithm and the driving algorithm to be compared in each target difference scenario.

[0057] The baseline driving algorithm and the driving algorithm to be compared can be different versions of autonomous driving algorithms. The target difference scenario refers to a simulated difference scenario for which functional positioning is required. The difference in planned trajectories between the baseline driving algorithm and the driving algorithm to be compared under the same simulated difference scenario satisfies a preset rule. This preset rule can be any rule used in the prior art for determining a difference scenario, and this application does not impose any specific limitation thereto.

[0058] The computer device may determine each target difference scenario and determine the raw performance difference between the baseline driving algorithm and the comparison driving algorithm, two different versions of the autonomous driving algorithm, in each target difference scenario. Compared to the baseline driving algorithm, the comparison driving algorithm generally adds one or more new features or iterates one or more functional features. The raw performance difference may reflect the degree of difference between the baseline driving algorithm and the comparison driving algorithm with each newly added and iterated feature enabled, in the corresponding target difference scenario.

[0059] It should be noted that the specific value of the raw difference performance can be determined based on a specific difference evaluation index, which may or may not include a trajectory difference index. This application does not impose any specific restrictions on this. For example, when the difference evaluation index is an acceleration difference index, the raw difference performance refers to the acceleration difference between different versions of the autonomous driving algorithm under the same target difference scenario.

[0060] S104: Acquire various pre-set functions to be located.

[0061] The functions to be located may be functions whose original performance differences corresponding to any target difference scenario need to be determined. These functions may be new functions added to the driving algorithm to be compared compared to the baseline driving algorithm, or functions that have undergone algorithmic iteration. Each function to be located may be pre-set by an engineer, and the number of functions to be located may be one or more, and this application does not impose any specific restrictions on this.

[0062] S106: For each of the functions to be located, only the function to be located of the driving algorithm to be compared is turned off, and each of the target difference scenarios is used to perform simulation tests on the driving algorithm to be compared after the function is turned off, so as to obtain the autonomous driving data of the function to be located in each of the target difference scenarios.

[0063] Specifically, for each pre-set function to be located, the computer device may disable the function to be located of the driving algorithm to be compared, while keeping the remaining functions enabled, to obtain a branch of the driving algorithm to be compared. The computer device may perform simulation tests on the branch of the driving algorithm to be compared using each target difference scenario to obtain autonomous driving data for the branch of the driving algorithm to be compared under each target difference scenario, that is, to obtain autonomous driving data corresponding to the function to be located. The autonomous driving data may record the behavioral state and / or driving state of the simulated vehicle controlled by the branch of the driving algorithm to be compared under each target difference scenario.

[0064] For example, the pre-set functions to be located are Function A and Function B, and the target difference scenarios are Scenario A and Scenario B. In this case, the computer device can disable Function A of the driving algorithm to be compared, and keep Function B and the remaining functions enabled. The computer device can then use Scenario A and Scenario B to simulate and test the driving algorithm to be compared with Function A disabled, to obtain the autonomous driving data corresponding to Function A. The computer device can also disable Function B of the driving algorithm to be compared, and keep Function A and the remaining functions enabled. The computer device can then use Scenario A and Scenario B to simulate and test the driving algorithm to be compared with Function B disabled, to obtain the autonomous driving data corresponding to Function B.

[0065] S108: For each of the target difference scenarios, based on the autonomous driving data of each of the functions to be located in the target difference scenario, determine the test difference performance corresponding to each of the functions to be located respectively, and based on each of the test difference performances and the original difference performance corresponding to the target difference scenario, determine the target function from each of the functions to be located, where the target function is the function that causes the original difference performance of the target difference scenario.

[0066] After S106, the computer can obtain autonomous driving data for each target-differentiation function under each target-differentiation scenario. In other words, for each target-differentiation scenario, it can correspond to autonomous driving data for each target-differentiation function. For each target-differentiation scenario, the computer device can determine the test difference performance corresponding to each target-differentiation function based on the autonomous driving data for each target-differentiation function under that target-differentiation scenario. This test difference performance can reflect the degree of difference between the target-differentiation driving algorithm and the baseline driving algorithm under the same target-differentiation scenario after the corresponding target-differentiation function in the target-differentiation driving algorithm is disabled.

[0067] For each target difference scenario, the computer device locates one or more functions that may cause the original difference performance among each function to be located as the target function based on the test difference performance corresponding to the target difference scenario and the original difference performance. In other words, there is a high probability that the original difference performance of the target difference scenario is caused by the one or more target functions.

[0068] In one embodiment, after determining the target functions corresponding to each target difference scenario, the computer device can bind each target difference scenario to its corresponding target function, and push the target difference scenario to the corresponding development engineer according to the corresponding target function, so that the development engineer can check in detail whether the function causes the target difference scenario.

[0069] In this application, computer equipment can automatically locate the differences between two different versions of autonomous driving algorithms to one or more specific functional features, so that engineers can perform manual verification based on the results of the positioning. In this way, the number of functions that require manual screening and positioning can be greatly reduced, thereby reducing the human resources consumed and improving testing efficiency.

[0070] In one embodiment, Figure 2 As shown, the step of determining each target difference scenario requiring functional positioning, and the original difference performance between the baseline driving algorithm and the driving algorithm to be compared in each target difference scenario, includes:

[0071] S202: Acquire each of the original difference scenarios, where the original difference scenarios are scenarios in which the difference in planned trajectories between the benchmark driving algorithm and the driving algorithm to be compared satisfies a preset rule.

[0072] The preset rules involved in this step can be any rules in the prior art for determining difference scenarios, and this application does not impose any specific restrictions on this.

[0073] In one embodiment, a computer device may simulate a benchmark driving algorithm using complete road segment data to obtain planned trajectory data output by the benchmark driving algorithm. Furthermore, the computer device may simulate a driving algorithm to be compared using the complete road segment data to obtain planned trajectory data output by the driving algorithm to be compared. Based on the planned trajectory data output by the benchmark driving algorithm and the planned trajectory data output by the driving algorithm to be compared, the computer device may calculate the difference in planned trajectories for each simulated frame, and based on this difference, select one or more original difference scenarios from the complete road segment data.

[0074] S204: Acquire various preset difference evaluation indicators.

[0075] The difference evaluation indicators can be indicators used to evaluate the differences in behavior and / or driving behavior between the baseline driving algorithm and the driving algorithm to be compared under the same simulation test scenario. The number and specific indicators of each difference evaluation indicator can be determined based on actual circumstances and are not specifically limited in this document.

[0076] S206: For each difference evaluation index, determine an initial difference performance for each original difference scenario under the difference evaluation index, which is used to reflect the degree of difference between the baseline driving algorithm and the driving algorithm to be compared; and based on each initial difference performance corresponding to the difference evaluation index, select an original difference scenario with a large degree of difference from each original difference scenario as a target difference scenario; and use the initial difference performance corresponding to each target difference scenario as the original difference performance corresponding to the target difference scenario.

[0077] Specifically, since the existence of trajectory planning differences does not necessarily mean that the driving algorithm to be compared has disadvantages that need to be resolved, after obtaining each original difference scenario, the computer device can separately determine the initial difference performance of each original difference scenario under each difference evaluation indicator. This initial difference performance can reflect the degree of difference between the baseline driving algorithm and the driving algorithm to be compared under the corresponding original difference scenario and the corresponding difference evaluation indicator.

[0078] For each difference evaluation indicator, the computer device can determine the degree of difference between the baseline driving algorithm and the driving algorithm to be compared under each original difference scenario and the difference evaluation indicator based on the initial difference performances corresponding to the difference evaluation indicator, and accordingly select the original difference scenarios with a large degree of difference under the difference evaluation indicator as the target difference scenarios corresponding to the difference evaluation indicator, and respectively use the initial difference performance corresponding to each target difference scenario as the original difference performance of the target difference scenario.

[0079] In this way, from the many original difference scenarios with different planning trajectories, the original difference scenarios that can truly reflect the problem can be screened out as the target difference scenarios, and subsequent steps can be executed, so as to achieve accurate comparison and reduce the subsequent calculation amount, thereby further improving test efficiency.

[0080] In one embodiment, each of the difference evaluation indicators includes an error difference indicator. That is, among the difference evaluation indicators, one of the difference evaluation indicators is an error difference indicator. The error difference indicator is an indicator used to evaluate the difference in system errors between the baseline driving algorithm and the driving algorithm to be compared under the same original difference scenario. Each of the target difference scenarios includes a first target difference scenario. That is, each target difference scenario includes the first target difference scenario, which is an original difference scenario in which a system error difference exists.

[0081] like Figure 3 As shown, the steps of respectively determining, for each of the original difference scenarios, an initial difference performance reflecting the degree of difference between the baseline driving algorithm and the driving algorithm to be compared under the difference evaluation index, and screening, based on each of the initial difference performances corresponding to the difference evaluation index, an original difference scenario with a large degree of difference from each of the original difference scenarios as a target difference scenario, include:

[0082] S302: Obtain first error information of the benchmark driving algorithm in each of the original difference scenarios respectively.

[0083] The first error message of the baseline driving algorithm in the original difference scenario can be used to indicate whether the baseline driving algorithm issued a system error message in the original difference scenario. Furthermore, if a problem or road condition that the baseline driving algorithm cannot safely resolve arises during the simulation, the baseline driving algorithm will issue a system error message.

[0084] S304: Obtain second error information of the driving algorithm to be compared in each of the original difference scenarios.

[0085] Similar to the first error message, the second error message of the driving algorithm to be compared in the original difference scenario may be used to reflect whether the driving algorithm to be compared issues a system error in the original difference scenario.

[0086] S306: Determine, based on the first error information and the second error information, the initial difference performance of each original difference scenario under the error difference indicator, wherein the initial difference performance is used to reflect whether there is an error difference between the baseline driving algorithm and the driving algorithm to be compared under the corresponding original difference scenario.

[0087] Specifically, for each original difference scenario, the computer device can determine, based on the first error information, whether the baseline driving algorithm issues a system error in that original difference scenario, and, based on the second error information, whether the driving algorithm to be compared issues a system error in that original difference scenario. Therefore, based on the first and second error information, the computer device can determine whether there is an error reporting difference between the baseline driving algorithm and the driving algorithm to be compared in that original difference scenario, i.e., determine the initial difference in performance of that original difference scenario under the error reporting difference metric. An error reporting difference refers to a situation where, for the same original difference scenario, one of the baseline driving algorithm and the driving algorithm to be compared issues a system error, while the other does not.

[0088] S308: Selecting each of the original difference scenarios with error reporting differences as the first target difference scenarios according to the initial difference performance of each of the original difference scenarios under the error reporting difference indicator.

[0089] After S306, the computer device can determine whether there is an error difference between the baseline driving algorithm and the driving algorithm to be compared in each original difference scenario based on the initial difference performance of each original difference scenario under the error difference index, and the computer device can declare the original difference scenario with the error difference as the first target difference scenario.

[0090] For example, the original difference scenarios include scenario C and scenario D. There is an error reporting difference between the baseline driving algorithm and the driving algorithm to be compared in scenario C, and there is no error reporting difference between the baseline driving algorithm and the driving algorithm to be compared in scenario D. The computer device can select scenario C as the first target difference scenario.

[0091] In this embodiment, by setting the error difference index in each difference evaluation index and selecting the original difference scene with error difference as the target difference scene, the original difference scene that can truly reflect the problem is screened out from multiple original difference scenes with planning trajectory differences as the target difference scene, thereby achieving accurate comparison and reducing subsequent calculations to further improve test efficiency.

[0092] In one embodiment, the step of determining the target function from each of the functions to be located based on each of the test difference performances and the original difference performance corresponding to the target difference scenario includes:

[0093] If the target difference scenario is the first target difference scenario, the test difference performance that is identical to the original difference performance corresponding to the first target difference scenario in each of the test difference performances is taken as the target test difference performance, and the function to be located corresponding to each of the target test difference performances is determined as the target function.

[0094] It will be appreciated that in S108, if the target difference scenario currently being processed is the first target difference scenario, the computer device may determine the test difference performance corresponding to each function to be positioned based on the autonomous driving data of each function to be positioned under the first target difference scenario. The test difference performance corresponding to each function to be positioned may be used to indicate whether there is a difference in error reporting between the baseline driving algorithm and the driving algorithm to be compared with the function to be positioned disabled under the first target difference scenario.

[0095] The computer device can use the test difference performance that is the same as the original difference performance of the first target difference scenario as the target test difference performance, and determine the target function that causes the original difference performance among the functions to be located according to the functions to be located corresponding to each target test difference performance.

[0096] Furthermore, the original differential performance of the first target differential scenario may also record the algorithm identifier of the autonomous driving algorithm that generated a system error and / or the algorithm identifier of the autonomous driving algorithm that did not generate a system error. Each test differential performance may also record the algorithm identifier of the autonomous driving algorithm that generated a system error and / or the algorithm identifier of the autonomous driving algorithm that did not generate a system error. The computer device may use the test differential performance among the test differential performances that has an error difference and generates a system algorithm that is consistent with the original differential performance as the target test differential performance, and determine the target function accordingly.

[0097] In this way, the target function that causes the error difference can be accurately located from all the functions to be located, thereby greatly reducing the number of functions that need to be manually screened and located, further reducing the human resources consumed, and improving testing efficiency.

[0098] In one embodiment, each of the difference evaluation indicators includes a simulation score difference indicator. That is, among the difference evaluation indicators, one of the difference evaluation indicators is a simulation score difference indicator. The simulation score difference indicator is an indicator used to evaluate the difference in simulation scores between a baseline driving algorithm and a driving algorithm to be compared under the same original difference scenario. In one embodiment, the simulation score difference indicator can be used to evaluate the difference in safety scores, comfort scores, and traffic congestion scores between different versions of autonomous driving algorithms under the same original difference scenario.

[0099] Each of the target difference scenarios includes a second target difference scenario, that is, each target difference scenario includes a second target difference scenario, and the second target difference scenario refers to an original difference scenario whose simulation score difference satisfies the scenario screening rule.

[0100] like Figure 4 As shown, the steps of respectively determining, for each of the original difference scenarios, an initial difference performance reflecting the degree of difference between the baseline driving algorithm and the driving algorithm to be compared under the difference evaluation index, and screening, based on each of the initial difference performances corresponding to the difference evaluation index, an original difference scenario with a large degree of difference from each of the original difference scenarios as a target difference scenario, include:

[0101] S402: Obtain a first simulation score of the benchmark driving algorithm in each of the original difference scenarios.

[0102] The first simulation score of the benchmark driving algorithm in the original difference scenario can be used to reflect the driving performance of the benchmark driving algorithm in the original difference scenario. In one embodiment, the first simulation score can include a first safety score, a first comfort score, and a first traffic congestion score.

[0103] S404: Obtain a second simulation score of the driving algorithm to be compared in each of the original difference scenarios.

[0104] Similar to the first simulation score, the second simulation score of the driving algorithm to be compared under the original difference scenario can be used to reflect the driving performance of the driving algorithm to be compared under the original difference scenario. In one embodiment, the second simulation score can include a second safety score, a second comfort score, and a second congestion score.

[0105] S406: For each of the original difference scenarios, calculate the difference between the first simulation score corresponding to the original difference scenario and the second simulation score corresponding to the original difference scenario, and use the difference as the initial difference performance corresponding to the original difference scenario under the simulation score difference indicator.

[0106] Specifically, for each original difference scenario, the computer device can calculate the difference between the two simulation scores based on the first simulation score of the baseline driving algorithm in the original difference scenario and the second simulation score of the driving algorithm to be compared in the original difference scenario, and use the calculated difference as the initial difference performance corresponding to the original difference scenario under the simulation score difference index.

[0107] Furthermore, if the simulation score difference indicator is used to evaluate the safety score difference, comfort score difference and congested traffic score difference of different versions of autonomous driving algorithms under the same original difference scenario, the first simulation score may include a first safety score, a first comfort score and a first congested traffic score, and the second simulation score may include a second safety score, a second comfort score and a second congested traffic score. Then, for each original difference scenario, the computer device may calculate a first difference between the first safety score and the second safety score corresponding to the original difference scenario, a second difference between the first comfort score and the second comfort score corresponding to the original difference scenario, and a third difference between the first congested traffic score and the second congested traffic score corresponding to the original difference scenario, and use the first difference, the second difference and the third difference as the initial difference performance corresponding to the original difference scenario.

[0108] S408: Sort the initial difference performances of each of the original difference scenarios under the simulation score difference index in descending order, and select the original difference scenarios corresponding to the first N initial difference performances as the second target difference scenarios, where N is a preset positive integer.

[0109] Specifically, for the initial difference performance corresponding to the original difference scenario under the simulation score difference index, the larger the initial difference performance value, the greater the degree of difference. With respect to the simulation score difference index, the computer device can sort the initial difference performances corresponding to the simulation evaluation difference index in descending order. The original difference scenarios corresponding to the first N (N total) initial difference performances after sorting are used as the second target difference scenarios.

[0110] Furthermore, if the simulation score difference index is used to evaluate the safety score difference, comfort score difference and traffic congestion score difference of different versions of autonomous driving algorithms under the same original difference scenario, the computer device can sort the first difference values ​​corresponding to each of the original difference scenarios in order from large to small, and select the original difference scenarios corresponding to the first N first difference values ​​after sorting as the second target difference scenarios; sort the second difference values ​​corresponding to each of the original difference scenarios in order from large to small, and select the original difference scenarios corresponding to the first N second difference values ​​after sorting as the second target difference scenarios; sort the third difference values ​​corresponding to each of the original difference scenarios in order from large to small, and select the original difference scenarios corresponding to the first N third differences after sorting as the second target difference scenarios.

[0111] In this way, the N original difference scenarios with the largest simulation score differences can be used as target difference scenarios, and subsequent steps can be performed, which can greatly reduce the number of functions that need to be manually screened and located, further reduce the human resources consumed, and improve test efficiency.

[0112] In one embodiment, each of the difference evaluation indicators includes a driving behavior difference indicator, i.e., one of the difference evaluation indicators is a driving behavior difference indicator. The driving behavior difference evaluation indicator is used to evaluate the difference in vehicle state between a simulated vehicle controlled by a baseline driving algorithm and a simulated vehicle controlled by a driving algorithm to be compared. Each of the target difference scenarios includes a third target difference scenario, which is an original difference scenario in which the driving behavior difference satisfies the scenario screening criteria.

[0113] like Figure 5 As shown, the steps of respectively determining, for each of the original difference scenarios, an initial difference performance reflecting the degree of difference between the baseline driving algorithm and the driving algorithm to be compared under the difference evaluation index, and screening, based on each of the initial difference performances corresponding to the difference evaluation index, an original difference scenario with a large degree of difference from each of the original difference scenarios as a target difference scenario, include:

[0114] S502: Obtain the first vehicle state of the benchmark driving algorithm in each of the original difference scenarios respectively.

[0115] In one embodiment, the first vehicle state may include the lighting status, driving trajectory, wheel angle status, whether entering automatic parking state, steering wheel status, horn honking status and brake change status of a simulated vehicle controlled by a benchmark driving algorithm in each original difference scenario.

[0116] S504: Obtain the second vehicle state of the driving algorithm to be compared in each of the original difference scenarios.

[0117] In one embodiment, the second vehicle state may include the lighting status, driving trajectory, wheel angle status, whether entering automatic parking state, steering wheel status, horn honking status and brake change status of the simulated vehicle controlled by the driving algorithm to be compared in each original difference scenario.

[0118] S506: For each of the original difference scenarios, a difference score is generated based on the first vehicle state corresponding to the original difference scenario and the second vehicle state corresponding to the original difference scenario, which is used to reflect the degree of difference in driving behavior between the baseline driving algorithm and the driving algorithm to be compared, and the difference score is used as the initial difference performance corresponding to the original difference scenario under the driving behavior difference index.

[0119] Specifically, for each original difference scenario, the computer device may integrate various vehicle conditions to generate a difference score that reflects the difference in driving behavior between the baseline driving algorithm and the driving algorithm to be compared. The difference score is positively correlated with the degree of difference in driving behavior. The computer device may use the calculated difference score as the initial difference performance corresponding to the original difference scenario under the driving behavior difference indicator.

[0120] S508: Sort the initial difference performances of each of the original difference scenarios under the driving behavior difference index in descending order, and select the original difference scenarios corresponding to the first N initial difference performances as the third target difference scenario, where N is a preset positive integer.

[0121] Specifically, for the initial difference performance corresponding to the original difference scenario under the driving behavior difference index, the larger the initial difference performance value, the greater the degree of difference. With respect to the driving behavior difference index, the computer device may sort the initial difference performances corresponding to the driving behavior difference index from largest to smallest. The original difference scenarios corresponding to the first N (of a total of N) initial difference performances after sorting are then used as the third target difference scenario.

[0122] In this way, the N original difference scenarios with the largest driving behavior differences can be used as target difference scenarios, and subsequent steps can be performed, which can greatly reduce the number of functions that need to be manually screened and located, further reduce the human resources consumed, and improve test efficiency.

[0123] In one embodiment, the step of determining the target function from each of the functions to be located based on each of the test difference performances and the original difference performance corresponding to the target difference scenario includes:

[0124] If the target difference scenario is the second target difference scenario, determining a screening interval according to the original difference performance corresponding to the second target difference scenario, the screening interval including the original difference performance;

[0125] If the target difference scenario is the third target difference scenario, determining a screening interval according to the original difference performance corresponding to the third target difference scenario, the screening interval including the original difference performance;

[0126] Target test difference performances falling within the screening interval are screened out from each of the test difference performances, and the target function is determined according to the to-be-located functions corresponding to each of the target test difference performances.

[0127] It will be appreciated that, in S108, if the currently processed target difference scenario is the second target difference scenario or the third target difference scenario, the computer device may determine the test difference performance corresponding to each function to be positioned based on the autonomous driving data of each function to be positioned under the currently processed target difference scenario. The test difference performance corresponding to each function to be positioned may be used to reflect the degree of difference in simulation scores between the baseline driving algorithm and the driving algorithm to be compared with the function to be positioned disabled under the second target difference scenario, or the degree of difference in driving behavior under the third target difference scenario.

[0128] If the target difference scenario currently being processed is the second target difference scenario or the third target difference scenario, the computer device may determine a screening interval based on the original difference performance corresponding to the second target difference scenario or the third target difference scenario. Furthermore, the screening interval may be a continuous interval including the original difference performance.

[0129] For each test difference performance corresponding to the target difference interval currently being processed, the computer device can use the test difference performance whose value falls within the screening interval as the target test difference performance, and determine the target function among each function to be located based on the functions to be located corresponding to each target test difference performance.

[0130] In this way, the target function that causes the difference in simulation scores and / or driving behavior can be accurately located from various functions to be located, thereby greatly reducing the number of functions that need to be manually screened and located, further reducing the human resources consumed, and improving test efficiency.

[0131] The function positioning device provided in an embodiment of the present application is described below. The function positioning device described below and the function positioning method described above can be referenced to each other.

[0132] In one embodiment, the present application provides a function positioning device 600. Figure 6 As shown, the device 600 includes a scene determination module 610 , a function acquisition module 620 , a simulation module 630 and a positioning module 640 .

[0133] in:

[0134] A scenario determination module 610 is configured to determine each target difference scenario for which function positioning is required, and the original difference performance between the baseline driving algorithm and the driving algorithm to be compared in each target difference scenario;

[0135] Function acquisition module 620, used to acquire various pre-set functions to be located;

[0136] a simulation module 630 configured to disable, for each of the functions to be located, only the function to be located of the driving algorithm to be compared, and perform simulation tests on the driving algorithm to be compared with the disabled function using each of the target difference scenarios, to obtain autonomous driving data for the function to be located under each of the target difference scenarios;

[0137] The positioning module 640 is used to determine, for each target difference scenario, the test difference performance corresponding to each of the functions to be positioned according to the autonomous driving data of each of the functions to be positioned in the target difference scenario, and determine a target function from each of the functions to be positioned based on each of the test difference performances and the original difference performance corresponding to the target difference scenario, where the target function is the function that causes the original difference performance of the target difference scenario.

[0138] In one embodiment, the scenario determination module 610 includes an original difference scenario acquisition unit, a difference evaluation index acquisition unit, and a scenario screening unit. The original difference scenario acquisition unit is used to acquire each original difference scenario, where the original difference scenario is a scenario in which the difference in the planned trajectory between the baseline driving algorithm and the driving algorithm to be compared satisfies a preset rule. The difference evaluation index acquisition unit is used to acquire each pre-set difference evaluation index. The scenario screening unit is used to determine, for each difference evaluation index, the initial difference performance of each original difference scenario under the difference evaluation index, which is used to reflect the degree of difference between the baseline driving algorithm and the driving algorithm to be compared, and based on the initial difference performance corresponding to the difference evaluation index, screen out original difference scenarios with a large degree of difference from each original difference scenario as target difference scenarios, and use the initial difference performance corresponding to each target difference scenario as the original difference performance corresponding to the target difference scenario.

[0139] In one embodiment, each of the difference evaluation indicators includes an error reporting difference indicator, and each of the target difference scenarios includes a first target difference scenario.

[0140] The scenario screening unit includes a first error information acquisition unit, a second error information acquisition unit, an error difference determination unit and a first selection unit. The first error information acquisition unit is used to respectively obtain the first error information of the baseline driving algorithm in each of the original difference scenarios. The second error information acquisition unit is used to respectively obtain the second error information of the driving algorithm to be compared in each of the original difference scenarios. The error difference determination unit is used to determine the initial difference performance of each of the original difference scenarios under the error difference index based on the first error information and the second error information. The initial difference performance is used to reflect whether there is an error difference between the baseline driving algorithm and the driving algorithm to be compared in the corresponding original difference scenario. The first selection unit is used to select each of the original difference scenarios with error differences as the first target difference scenario based on the initial difference performance of each of the original difference scenarios under the error difference index.

[0141] In one embodiment, the positioning module 640 includes a first positioning unit. The first positioning unit is configured to, when the target difference scenario is the first target difference scenario, use, among the test difference performances, a test difference performance that is identical to an original difference performance corresponding to the first target difference scenario as a target test difference performance, and determine the to-be-positioned function corresponding to each target test difference performance as the target function.

[0142] In one embodiment, each of the difference evaluation indicators includes a simulation score difference indicator, and each of the target difference scenarios includes a second target difference scenario.

[0143] The scenario screening unit includes a first simulation score acquisition unit, a second simulation score acquisition unit, a simulation score difference determination unit and a second selection unit. The first simulation score acquisition unit is used to respectively obtain the first simulation score of the benchmark driving algorithm in each of the original difference scenarios. The second simulation score acquisition unit is used to respectively obtain the second simulation score of the driving algorithm to be compared in each of the original difference scenarios. The simulation score difference determination unit is used to calculate the difference between the first simulation score corresponding to the original difference scenario and the second simulation score corresponding to the original difference scenario for each of the original difference scenarios, and use the difference as the initial difference performance corresponding to the original difference scenario under the simulation score difference index. The second selection unit is used to sort the initial difference performance of each of the original difference scenarios under the simulation score difference index in descending order, and select the original difference scenarios corresponding to the first N initial difference performances as the second target difference scenarios, wherein N is a preset positive integer.

[0144] In one embodiment, each of the difference evaluation indicators includes a driving behavior difference indicator, and each of the target difference scenarios includes a third target difference scenario.

[0145] The scenario screening unit includes a first vehicle state acquisition unit, a second vehicle state acquisition unit, a state difference determination unit, and a third selection unit. The first vehicle state acquisition unit is configured to respectively acquire a first vehicle state of the baseline driving algorithm under each of the original difference scenarios. The second vehicle state acquisition unit is configured to respectively acquire a second vehicle state of the driving algorithm to be compared under each of the original difference scenarios. The state difference determination unit is configured to generate, for each of the original difference scenarios, a difference score reflecting the degree of difference in driving behavior between the baseline driving algorithm and the driving algorithm to be compared based on the first vehicle state corresponding to the original difference scenario and the second vehicle state corresponding to the original difference scenario, and to use the difference score as the initial difference performance of the original difference scenario under the driving behavior difference index. The third selection unit is configured to sort the initial difference performance of each of the original difference scenarios under the driving behavior difference index in descending order, and select the original difference scenarios corresponding to the first N initial difference performances as the third target difference scenario, where N is a preset positive integer.

[0146] In one embodiment, the positioning module 640 includes a first interval determination unit, a second interval determination unit, and a second positioning unit. The first interval determination unit is used to determine a screening interval based on the original difference performance corresponding to the second target difference scenario if the target difference scenario is the second target difference scenario, and the screening interval includes the original difference performance. The second interval determination unit is used to determine a screening interval based on the original difference performance corresponding to the third target difference scenario if the target difference scenario is the third target difference scenario, and the screening interval includes the original difference performance. The second positioning unit is used to screen out the target test difference performances that fall within the screening interval from each of the test difference performances, and determine the target function based on the to-be-positioned function corresponding to each of the target test difference performances.

[0147] In one embodiment, the present application further provides a storage medium storing computer-readable instructions. When the computer-readable instructions are executed by one or more processors, the one or more processors execute the steps of the function positioning method described in any of the above embodiments.

[0148] In one embodiment, the present application further provides a computer device having computer-readable instructions stored therein, which, when executed by one or more processors, causes the one or more processors to perform the steps of the function location method described in any of the above embodiments.

[0149] Schematically, Figure 7 This is a schematic diagram of the internal structure of a computer device provided in an embodiment of the present application. In one example, the computer device may be a server. Figure 7 Computer device 900 includes a processing component 902, which further includes one or more processors, and memory resources represented by memory 901 for storing instructions executable by processing component 902, such as application programs. The application programs stored in memory 901 may include one or more modules, each corresponding to a set of instructions. Furthermore, processing component 902 is configured to execute the instructions to perform the steps of the function location method described in any of the above embodiments.

[0150] The computer device 900 may further include a power supply component 903 configured to perform power management of the computer device 900, a wired or wireless network interface 904 configured to connect the computer device 900 to a network, and an input / output (I / O) interface 905. The computer device 900 may operate based on an operating system stored in the memory 901, such as Windows Server™, Mac OS X™, Unix™, Linux™, Free BSD™, or the like.

[0151] Those skilled in the art will understand that the internal structure of the computer device shown in the present application is merely a block diagram of a partial structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0152] Finally, it should be noted that, in this article, relational terms such as first and second are merely used to distinguish one entity or operation from another, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprise," "include," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. Without further restriction, an element defined by the phrase "comprising a..." does not exclude the presence of other identical elements in the process, method, article, or device comprising the element. Herein, "one," "said," "the," and "its" may also include plural forms unless the context clearly indicates otherwise. A plurality refers to at least two, such as 2, 3, 5, or 8. "And / or" includes any and all combinations of the relevant listed items.

[0153] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The various embodiments can be combined as needed, and the same or similar parts can be referenced to each other.

[0154] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present application. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application is not limited to the embodiments shown herein, but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A function positioning method, characterized in that: The method comprises: Determining each target difference scenario for which functional positioning is required, and the original difference performance between the baseline driving algorithm and the driving algorithm to be compared in each target difference scenario; Get the preset functions to be located; For each of the functions to be located, only the function to be located of the driving algorithm to be compared is disabled, and simulation tests are performed on the driving algorithm to be compared with the disabled function using each of the target difference scenarios, so as to obtain autonomous driving data of the function to be located in each of the target difference scenarios; For each target difference scenario, determining the test difference performance corresponding to each function to be located based on the autonomous driving data of each function to be located in the target difference scenario, and determining a target function from each function to be located based on each test difference performance and the original difference performance corresponding to the target difference scenario, where the target function is the function that causes the original difference performance of the target difference scenario; The step of determining each target difference scenario requiring functional positioning, and the original difference performance between the baseline driving algorithm and the driving algorithm to be compared in each target difference scenario, includes: Obtaining each original difference scenario, where the original difference scenario is a scenario in which the difference in planned trajectories between the baseline driving algorithm and the driving algorithm to be compared satisfies a preset rule; Obtain various pre-set difference evaluation indicators; For each of the difference evaluation indicators, an initial difference performance for each of the original difference scenarios under the difference evaluation indicator, which is used to reflect the degree of difference between the baseline driving algorithm and the driving algorithm to be compared, is determined. Based on the initial difference performances corresponding to the difference evaluation indicator, original difference scenarios with a large degree of difference are screened out from the original difference scenarios as target difference scenarios, and the initial difference performance corresponding to each target difference scenario is used as the original difference performance corresponding to the target difference scenario.

2. The function positioning method according to claim 1, characterized in that: Each of the difference evaluation indicators includes an error reporting difference indicator, and each of the target difference scenarios includes a first target difference scenario; The step of determining, for each of the original difference scenarios, an initial difference performance reflecting the degree of difference between the baseline driving algorithm and the driving algorithm to be compared under the difference evaluation index, and screening, based on each of the initial difference performances corresponding to the difference evaluation index, an original difference scenario with a large degree of difference from each of the original difference scenarios as a target difference scenario, includes: respectively obtaining first error information of the benchmark driving algorithm in each of the original difference scenarios; Respectively obtaining second error information of the driving algorithm to be compared in each of the original difference scenarios; determining, based on the first error reporting information and the second error reporting information, an initial difference performance under the error reporting difference indicator for each of the original difference scenarios, the initial difference performance being used to reflect whether there is an error reporting difference between the baseline driving algorithm and the driving algorithm to be compared under the corresponding original difference scenario; According to the initial difference performance of each of the original difference scenes under the error reporting difference indicator, each of the original difference scenes with error reporting differences is selected as the first target difference scene.

3. The function positioning method according to claim 2, characterized in that: The step of determining the target function from each of the functions to be located based on each of the test difference performances and the original difference performance corresponding to the target difference scenario includes: If the target difference scenario is the first target difference scenario, the test difference performance that is identical to the original difference performance corresponding to the first target difference scenario in each of the test difference performances is taken as the target test difference performance, and the function to be located corresponding to each of the target test difference performances is determined as the target function.

4. The function positioning method according to claim 1, characterized in that: Each of the difference evaluation indicators includes a simulation score difference indicator, and each of the target difference scenarios includes a second target difference scenario; The step of determining, for each of the original difference scenarios, an initial difference performance reflecting the degree of difference between the baseline driving algorithm and the driving algorithm to be compared under the difference evaluation index, and screening, based on each of the initial difference performances corresponding to the difference evaluation index, an original difference scenario with a large degree of difference from each of the original difference scenarios as a target difference scenario, includes: respectively obtaining a first simulation score of the benchmark driving algorithm in each of the original difference scenarios; respectively obtaining a second simulation score of the driving algorithm to be compared under each of the original difference scenarios; For each of the original difference scenarios, calculating the difference between the first simulation score corresponding to the original difference scenario and the second simulation score corresponding to the original difference scenario, and using the difference as the initial difference performance corresponding to the original difference scenario under the simulation score difference indicator; The initial difference performances of each of the original difference scenarios under the simulation score difference index are sorted in order from large to small, and the original difference scenarios corresponding to the first N initial difference performances are selected as the second target difference scenarios, where N is a preset positive integer.

5. The function positioning method according to claim 4, characterized in that: The step of determining the target function from each of the functions to be located based on each of the test difference performances and the original difference performance corresponding to the target difference scenario includes: If the target difference scenario is the second target difference scenario, determining a screening interval according to the original difference performance corresponding to the second target difference scenario, the screening interval including the original difference performance; Target test difference performances falling within the screening interval are screened out from each of the test difference performances, and the target function is determined according to the to-be-located functions corresponding to each of the target test difference performances.

6. The function positioning method according to claim 1, characterized in that: Each of the difference evaluation indicators includes a driving behavior difference indicator, and each of the target difference scenarios includes a third target difference scenario; The step of determining, for each of the original difference scenarios, an initial difference performance reflecting the degree of difference between the baseline driving algorithm and the driving algorithm to be compared under the difference evaluation index, and screening, based on each of the initial difference performances corresponding to the difference evaluation index, an original difference scenario with a large degree of difference from each of the original difference scenarios as a target difference scenario, includes: respectively obtaining a first vehicle state of the benchmark driving algorithm in each of the original difference scenarios; respectively obtaining a second vehicle state of the driving algorithm to be compared under each of the original difference scenarios; For each of the original difference scenarios, generating a difference score reflecting the degree of difference in driving behavior between the baseline driving algorithm and the driving algorithm to be compared based on a first vehicle state corresponding to the original difference scenario and a second vehicle state corresponding to the original difference scenario, and using the difference score as an initial difference performance corresponding to the original difference scenario under the driving behavior difference index; The initial difference performances of each of the original difference scenarios under the driving behavior difference index are sorted in order from large to small, and the original difference scenarios corresponding to the first N initial difference performances are selected as the third target difference scenarios, where N is a preset positive integer.

7. The function positioning method according to claim 6, characterized in that: The step of determining the target function from each of the functions to be located based on each of the test difference performances and the original difference performance corresponding to the target difference scenario includes: If the target difference scenario is the third target difference scenario, determining a screening interval according to the original difference performance corresponding to the third target difference scenario, the screening interval including the original difference performance; Target test difference performances falling within the screening interval are screened out from each of the test difference performances, and the target function is determined according to the to-be-located functions corresponding to each of the target test difference performances.

8. A function positioning device, characterized in that: The device comprises: A scenario determination module is used to determine each target difference scenario for which function positioning is required, and the original difference performance between the baseline driving algorithm and the driving algorithm to be compared in each target difference scenario; Function acquisition module, used to obtain each pre-set function to be located; a simulation module configured to, for each of the functions to be located, disable only the function to be located of the driving algorithm to be compared, and perform simulation tests on the driving algorithm to be compared after the function is disabled using each of the target difference scenarios, so as to obtain autonomous driving data of the function to be located in each of the target difference scenarios; a positioning module for determining, for each target difference scenario, a test difference performance corresponding to each function to be positioned based on the autonomous driving data of each function to be positioned in the target difference scenario, and determining a target function from each function to be positioned based on each test difference performance and an original difference performance corresponding to the target difference scenario, where the target function is the function that causes the original difference performance of the target difference scenario; The scene determination module includes: an original difference scenario acquisition unit, configured to acquire each original difference scenario, wherein the original difference scenario is a scenario in which the difference in planned trajectories between the baseline driving algorithm and the driving algorithm to be compared satisfies a preset rule; A difference evaluation index acquisition unit, used to acquire various preset difference evaluation indicators; a scenario screening unit for determining, for each difference evaluation index, an initial difference performance for each original difference scenario under the difference evaluation index, which is used to reflect the degree of difference between the baseline driving algorithm and the driving algorithm to be compared; and screening, based on the initial difference performance corresponding to the difference evaluation index, original difference scenarios with a large degree of difference from the original difference scenarios as target difference scenarios; and using the initial difference performance corresponding to each target difference scenario as the original difference performance corresponding to the target difference scenario.

9. A storage medium, characterized in that: The storage medium stores computer-readable instructions, which, when executed by one or more processors, enable the one or more processors to perform the steps of the function positioning method according to any one of claims 1 to 7.

10. A computer device, characterized in that: include: one or more processors, and memory; The memory stores computer-readable instructions, and when the computer-readable instructions are executed by the one or more processors, the steps of the function positioning method according to any one of claims 1 to 7 are performed.

Images