Automobile simulation test scene generation method and device, terminal and medium

Abstract

The invention provides an automobile simulation test scene generation method and device, a terminal and a medium. The method comprises the steps that a multi-level scene element structure model composed of a plurality of scene elements, element parameter values, the relative importance of the scene elements in the level, test combination dimensions and a test constraint condition set are acquired; according to the relative importance and the scene element structure model, obtaining a comprehensive importance weight; processing the scene element structure model, the scene elements, the element parameter values, the comprehensive importance values, the test combination dimensions and the test constraint condition set through a pairwise combination test algorithm to generate a simulation test case set; and performing scene simulation on the simulation test case set to generate a simulation test scene. The connotation of the test scene can be deconstructed, and the scene element comprehensive importance is introduced to improve the traditional test case generation algorithm, so that the test scene with high coverage and strong practicability can be quickly and directionally generated.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of automatic driving simulation, in particular to a method and device for generating a simulation test scene for a vehicle, a terminal and a medium. BACKGROUND

[0002] With the wide application of automatic driving technology, in order to verify the function and performance of the automatic driving system, improve the safety and reliability of the automatic driving vehicle, accelerate the research and development and listing time of the automatic driving vehicle, and at the same time improve the public's confidence in the automatic driving vehicle, a large number of simulation tests need to be carried out on the automatic driving vehicle during the product development stage. Therefore, how to obtain a simulation test environment that can simulate various situations in a real road environment, simulate different traffic situations, and be as close as possible to the situations in the real road environment is particularly important.

[0003] At present, the construction of intelligent networked vehicle simulation test scene is mainly through manual construction and automatic generation by traditional algorithm. However, manual construction of simulation test scene has problems such as high cost and low efficiency, while automatic generation of simulation test scene by traditional algorithm solves the problem of manual construction, but has problems such as low coverage of test scene and insufficient practicability, and cannot be well applied to intelligent networked vehicle simulation test. SUMMARY

[0004] In view of the deficiencies in the prior art, the present application provides a method for generating a simulation test scene for a vehicle, which aims to quickly generate a test scene with high coverage and strong practicability, thereby supporting the testing and verification of intelligent networked vehicles.

[0005] In a first aspect, the present application provides a method for generating a simulation test scene for a vehicle, comprising: obtaining a multi-level scene element structure model composed of a plurality of scene elements, and obtaining element parameter values of the scene elements, relative importance of the scene elements in the level, test combination dimension, and a set of test constraints; obtaining a comprehensive importance weight of the scene elements according to the relative importance and the scene element structure model; processing the scene element structure model, the scene elements, the element parameter values, the comprehensive importance value, the test combination dimension, and the set of test constraints by a pair-wise combination test algorithm to generate a simulation test case set; performing scene simulation on the simulation test case set to generate a simulation test scene.

[0006] Optionally, the step of processing the scene element structure model, the scene elements, the comprehensive importance value of the element parameters, the test combination dimension, and the test constraint set using a pairwise combination testing algorithm to generate a simulation test case set includes: Based on the test combination dimension, the scene elements in the scene element structure model are paired up in pairs to calculate possible element combinations. Based on the set of constraints, invalid combinations are eliminated from the possible combinations of elements to obtain valid pairwise combinations; Mark the valid pairwise combinations as uncovered to obtain a list of uncovered combinations; The simulation test case set is obtained by filtering the candidate test cases based on the uncovered combination list, the scene element structure model, the comprehensive importance value, the element parameter values ​​of the scene elements, and the uncovered combination list.

[0007] Optionally, the step of filtering the candidate test cases based on the uncovered combination list, the scene element structure model, the comprehensive importance value, the element parameter values ​​of the scene elements, and the uncovered combination list to obtain a simulation test case set includes: Generate a test suite; Compare the test set with the list of uncovered combinations; If no pairwise combination in the test set is fully covered by all pairwise combinations in the uncovered combination list, then all possible candidate test cases under the constraint condition set are generated by the Cartesian product of the scene element and the element parameter value. The complexity of the test cases is obtained based on the scene element structure model, the element parameter values ​​of the scene elements, and the comprehensive importance value. Select the candidate test cases with higher test case complexity and coverage of more uncovered combinations among the element parameter values ​​as the optimal test cases; Add the optimal test case to the test set, and compare the test set with the list of uncovered combinations; If every pair of combinations in the test set is completely covered by every pair of combinations in the list of uncovered combinations, then the test set will be used as a set of simulation test cases. If no pairwise combination in the test set is completely covered by all pairwise combinations in the uncovered combination list, a loop is executed to generate all possible candidate test cases under the constraint condition set through the Cartesian product of the scene element and the element parameter value.

[0008] Optionally, the candidate test cases with higher test case complexity and covering more uncovered combinations among the selected element parameter values ​​are selected as the optimal test cases, including: ; in, For the first The composite preference value of each candidate test case; For the first The complexity of each test case; The maximum complexity among the candidate test cases; The minimum complexity among the candidate test cases; For the first The number of candidate test cases that cover the number of uncovered combinations; This represents the maximum number of uncovered / uncovered combinations among the candidate test cases; This represents the minimum number of uncovered / covered combinations among the candidate test cases. This is the weighting factor.

[0009] Optionally, based on the relative importance and the scene element structure model, the comprehensive importance weight of the scene elements is obtained, including: Based on the relative importance of the scene elements in the scene element structure model at their respective levels, the relative importance matrix information is obtained. Calculate the eigenvector corresponding to the largest eigenvalue of the relative importance matrix at each level in the relative importance matrix information; The relative importance weights are obtained by normalizing the feature vectors. The overall importance weight of scene elements is obtained based on the relative importance weight and the overall importance weight of the previous layer.

[0010] Optionally, before obtaining a multi-level scene element structure model composed of several scene elements, and before obtaining the element parameter values ​​of the scene elements, the relative importance of the scene elements in their respective levels, the test combination dimension, and the set of test constraints, the process includes: Obtain the functional requirements for intelligent connected vehicle simulation testing; Based on the functional requirements and the pre-set functional requirements database, a multi-level scene element structure model consisting of several scene elements is obtained. Display the scene element structure model; Receive the element parameter values ​​of the scene elements returned by the displayed scene element structure model, the relative importance of the scene elements at their respective levels, the test combination dimension, and the set of test constraints.

[0011] Optionally, based on functional requirements and a pre-defined functional requirements database, a multi-level scene element structure model consisting of several scene elements is obtained, including: Semantic analysis is performed on the aforementioned functional requirements to obtain requirement scenario information; The required scenario information is matched and analyzed with a preset functional requirement database from two dimensions: traffic environment and test vehicle, to obtain a multi-level scenario element structure model composed of several scenario elements.

[0012] Secondly, embodiments of this application provide an automotive simulation test scenario generation device, comprising: The parameter acquisition module is used to acquire a multi-level scene element structure model composed of several scene elements, as well as to acquire the element parameter values ​​of the scene elements, the relative importance of the scene elements in their respective levels, the test combination dimension, and the set of test constraints. The importance determination module is used to obtain the comprehensive importance weight of scene elements based on the relative importance and the scene element structure model; The test case set determination module is used to process the scene element structure model, the scene elements, the element parameter values, the comprehensive importance value, the test combination dimension, and the test constraint set through a pairwise combination test algorithm to generate a simulation test case set; The scenario simulation module is used to perform scenario simulation on the simulation test case set and generate simulation test scenarios.

[0013] Thirdly, embodiments of this application provide a terminal device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the vehicle simulation test scenario generation method as described in any one of the first aspects above.

[0014] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the vehicle simulation test scenario generation method as described in any one of the first aspects above.

[0015] Fifthly, embodiments of this application provide a computer program product that, when run on a terminal device, causes the terminal device to execute the automotive simulation test scenario generation method described in any one of the first aspects.

[0016] In this embodiment, a multi-level scene element structure model composed of several scene elements is obtained, along with the element parameter values, relative importance of the scene elements at their respective levels, test combination dimension, and test constraint set. Based on the relative importance and the scene element structure model, a comprehensive importance weight for each scene element is obtained. A pairwise combination testing algorithm is used to process the scene element structure model, the scene elements, the element parameter values, the comprehensive importance value, the test combination dimension, and the test constraint set to generate a simulation test case set. Scene simulation is then performed on the simulation test case set to generate a simulation test scenario. By deconstructing the connotation of the test scenario and introducing a comprehensive importance weight for scene elements to improve traditional test case generation algorithms, highly comprehensive and practical test scenarios can be quickly generated, thereby supporting the testing and verification of intelligent connected vehicles. Attached Figure Description

[0017] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the accompanying drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn to scale.

[0018] Figure 1 This is a flowchart illustrating a method for generating an automotive simulation test scenario according to an embodiment of this application; Figure 2 This is a schematic diagram of the intelligent connected vehicle simulation test function requirements table of the vehicle simulation test scenario generation method provided in this application; Figure 3 This is a flowchart illustrating the second embodiment of the vehicle simulation test scenario generation method provided in this application; Figure 4 This is a flowchart of the improved PICT algorithm for generating test cases in the automotive simulation test scenario generation method provided in this application; Figure 5 This is a schematic diagram of the hierarchical structure of test scene elements in the automotive simulation test scene generation method provided in this application; Figure 6 This is a schematic diagram of some elements of the test case for the scenario of turning right through a T-junction without traffic lights in the vehicle simulation test scenario generation method provided in this application; Figure 7 This is a schematic diagram of a scene segment of a right turn through a T-junction without traffic lights, using the vehicle simulation test scene generation method provided in this application; Figure 8 This is a schematic diagram of the structure of the vehicle simulation test scene generation device provided in the embodiments of this application; Figure 9 This is a schematic diagram of the structure of the terminal device provided in the embodiments of this application. Detailed Implementation

[0019] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0020] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0021] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0022] The vehicle simulation test scenario generation method provided in this application can be executed by a vehicle simulation test scenario generation device. The device generates a vehicle simulation test scenario by acquiring several scene elements, a multi-level scene element structure model, and obtaining the element parameter values, relative importance of the scene elements at their respective levels, test combination dimensions, and a set of test constraints. Based on the relative importance and the scene element structure model, a comprehensive importance weight for the scene elements is obtained. A pairwise combination test algorithm is used to process the scene elements, element parameter values, comprehensive importance values, test combination dimensions, and test constraint sets to generate a simulation test case set. Scene simulation is then performed on the simulation test case set to generate a simulation test scenario.

[0023] Figure 1 This illustration shows a schematic flowchart of the vehicle simulation test scenario generation method provided in an embodiment of this application. It is provided as an example and not as a limitation. This method can be applied to the aforementioned vehicle simulation test scenario generation device, or it can be a method for users or operators to operate and make judgments on the vehicle simulation test scenario generation device. Figure 1 As shown, the method may include: S10, obtain a multi-level scene element structure model composed of several scene elements, and obtain the element parameter values ​​of the scene elements, the relative importance of the scene elements in their respective levels, the test combination dimension, and the set of test constraints. In order to quickly generate test scenarios with high coverage and strong practicality, the automotive simulation test scenario generation device acquires a multi-level scenario element structure model composed of several scenario elements, as well as the element parameter values ​​of the scenario elements, the relative importance of the scenario elements in their respective levels, the test combination dimension, and the set of test constraints.

[0024] The scene element structure model is a multi-level structured model composed of several scene elements. It can be input by the operator or automatically generated based on the operator's input of test requirements. The structure of the scene element structure model can be as follows: Figure 2 As shown. The test combination dimension N can be 2 or 3. Test constraint set. A set of one or more constraints; for example, constraints could be "the vehicle speed shall not exceed 80 km / h in rainy or foggy weather", "the vehicle speed on highways shall not exceed 120 km / h", and "the driving speed on rural roads shall not exceed 60 km / h on curves".

[0025] The scene element structure model can be expressed as: ; in, This refers to the j-th scene element in the i-th layer. for Subscript The set that constitutes; The number of levels in the architecture; This represents the number of scene elements in the i-th layer of the architecture. Scene elements are those that cannot be further subdivided into sub-elements. Test cases for a scene can be generated by combining the parameter values ​​of different scene elements.

[0026] in, The relative importance of scene elements at their respective levels. That is, for the scene elements in the i-th layer and The relative importance of (where) , , (Number of scene elements in the i-th layer), relative importance The value of can be obtained through scaling and the Delphi method (expert scoring method).

[0027] Among them, scene element categories Several scene elements Composed of various elements; each scene element It contains multiple feature parameter values. Composition. Scene element categories and scene elements Relationships between elements of a scene and the values ​​of its element parameters The relationship between them is as follows: ; ; in, ; Subscript; Scene elements The set of possible values ​​for a feature parameter. For example... If the parameter can take 3 values, then ={1, 2, 3}.

[0028] As one implementation method, before acquiring a multi-level scene element structure model composed of several scene elements, and before acquiring the element parameter values ​​of the scene elements, the relative importance of the scene elements at their respective levels, the test combination dimension, and the set of test constraints, the method includes: acquiring the functional requirements for intelligent connected vehicle simulation testing; obtaining a multi-level scene element structure model composed of several scene elements based on the functional requirements and a preset functional requirement database; displaying the scene element structure model; and receiving the element parameter values ​​of the scene elements, the relative importance of the scene elements at their respective levels, the test combination dimension, and the set of test constraints returned based on the displayed scene element structure model.

[0029] Before acquiring a multi-level scene element structure model composed of several scene elements and the relative importance of scene elements at their respective levels, the automotive simulation test scenario generation device can obtain the functional requirements for intelligent connected vehicle simulation testing. Based on these functional requirements and a pre-set functional requirement database, it generates a multi-level scene element structure model composed of several scene elements. This model is then displayed on a monitor so that operators can easily see it. The operator then inputs the element parameter values, the relative importance of each scene element at its level, the test combination dimension, and the set of test constraints. The automotive simulation test scenario generation device receives these parameters returned from the displayed scene element structure model.

[0030] As one implementation method, based on functional requirements and a preset functional requirements database, a multi-level scene element structure model consisting of several scene elements is obtained. This may include: performing semantic analysis on the functional requirements to obtain requirement scene information; and matching and parsing the requirement scene information with the preset functional requirements database from two dimensions: traffic environment and test vehicle, to obtain a multi-level scene element structure model consisting of several scene elements.

[0031] This approach allows for the targeted identification and understanding of elements within specific scenarios based on testing functional requirements, better capturing the characteristics and relationships of the scenario. This lays the foundation for more efficient and automated generation of test scenarios. Because scenario elements are diverse, and their parameter values ​​vary significantly across different scenarios, directly deconstructing a scenario without first analyzing it based on testing requirements would result in an overwhelming set of scenarios, leading to dimensionality explosion and hindering subsequent automated scenario generation. Analyzing testing functional requirements allows for the categorization of complex and diverse scenarios into limited types, reducing complexity and improving efficiency. In short, it generates useful scenarios based on requirements, rather than generating irrelevant ones. For example, if the functional requirement is "generate a main road overtaking test scenario," the scenario information should focus on this main road overtaking test scenario, rather than test scenarios on highways or secondary roads.

[0032] Before deconstructing the scenarios, it is necessary to analyze the functional requirements of intelligent connected vehicle simulation testing. By analyzing these requirements, the complex and diverse scenarios can be divided into a limited number of categories. This allows for the targeted identification and generation of scenarios that meet the testing needs, reducing complexity and improving efficiency. Currently, the specific functional testing requirements for intelligent connected vehicle simulation testing are as follows: Figure 2 As shown, since the scenarios and their elements are inexhaustible, this table only lists a portion. The pre-defined functional requirements database can be found by referring to... Figure 2 As shown, the preset functional requirements database is much larger and contains more scenario elements than the specific functional test requirements.

[0033] Based on specific requirements and scenario information, the scenario is deconstructed using the Unified Modeling Language (UML) from two dimensions: traffic environment elements (meteorological elements, static road information, dynamic road information, and traffic participant information) and test vehicle elements (test vehicle basic elements, test vehicle target information, and test vehicle driving behavior). Meteorological elements include weather and lighting; static road information includes lane information, traffic facilities, and obstacles; dynamic road information includes dynamic changes in roads and road facilities; and traffic participant information includes pedestrian, non-motorized vehicle, and motorized vehicle information. Test vehicle basic elements include geometric features, performance characteristics, and driving systems; test vehicle target information includes perception and recognition functions and path planning functions; and test vehicle driving behavior includes lateral and longitudinal movement behaviors.

[0034] S20, Based on the relative importance and the scene element structure model, obtain the comprehensive importance weight of the scene elements; After acquiring a multi-level scene element structure model consisting of several scene elements and the relative importance of the scene elements at their respective levels, the automotive simulation test scene generation device obtains the comprehensive importance weight of the scene elements based on the relative importance and the scene element structure model.

[0035] As one implementation method, obtaining the comprehensive importance weight of scene elements based on the relative importance and the scene element structure model may include: obtaining relative importance matrix information based on the relative importance of the scene element at its level in the scene element structure model; calculating the eigenvector corresponding to the largest eigenvalue of the relative importance matrix at each level in the relative importance matrix information; normalizing the eigenvector to obtain the relative importance weight; and obtaining the comprehensive importance weight of the scene element based on the relative importance weight and the comprehensive importance weight value of the previous level.

[0036] In other words, the automotive simulation test scenario generation device acquires... and Then, based on the relative importance of each scene element in its respective layer and the scene element structure model, the judgment matrix for each layer is first constructed, as follows: ; In the formula Let be the judgment matrix for each scene element in the i-th layer. (in , , (The number of scene elements in the i-th layer) represents the number of scene elements in the i-th layer. and The relative importance of can be obtained through scaling and the Delphi method (expert scoring method).

[0037] Then, calculate the judgment matrix. Maximum eigenvalue The corresponding eigenvectors are obtained by normalizing the eigenvectors through numerical summation. That is, to obtain the relative importance weights of scene elements. The relative importance weights of scene elements It can be represented as: ; In the formula, , For the i-th layer scene, the first... The relative importance weights of each scene element.

[0038] Since the weights of scene elements at different levels are not based on the same reference system, the importance of scene element classes is affected by the importance of scene elements in the previous level. To obtain a consistent description of the importance of elements to the scene, a comprehensive importance index is introduced to calculate the comprehensive importance weight of each scene element. The overall importance weight of scene elements It can be represented as: ; ; In the formula, This represents the comprehensive importance weight matrix of the scene elements in the i-th layer. In the i-th layer scene, the first... The overall importance weight of each scene element, In the i-th layer scene, the first... The overall importance weight value of the parent scene element at the next higher level (i-1 level) of each scene element.

[0039] S30, the scene element structure model, the scene elements, the element parameter values, the comprehensive importance value, the test combination dimension, and the test constraint set are processed by the pairwise combination test algorithm to generate a simulation test case set; The automotive simulation test scenario generation device, after obtaining the comprehensive importance weights of scenario elements and acquiring a multi-level scenario element structure model composed of several scenario elements, and obtaining the element parameter values, test combination dimensions, and test constraint sets of the scenario elements, processes the scenario element structure model, the scenario elements, the element parameter values, the comprehensive importance values, the test combination dimensions, and the test constraint set using a pairwise combination test algorithm (PICT algorithm) to generate a simulation test case set. The traditional Pairwise Combination Test (PICT) algorithm can generate test cases with good coverage and minimize the number of test cases, but the traditional PPICT algorithm uses random selection when generating test cases, which may result in low scenario complexity. Therefore, to improve scenario complexity while ensuring coverage and scenario diversity, the traditional PPICT algorithm is improved by introducing a comprehensive importance weight for scenario elements. The improved PPICT algorithm is as follows: Figure 4 As shown.

[0040] Furthermore, referring to Figure 3 , Figure 3 This is a flowchart illustrating the second embodiment of the vehicle simulation test scenario generation method of the present invention. Based on the above... Figure 3 The illustrated embodiment uses a pairwise combination testing algorithm to process the scene element structure model, the scene elements, the element parameter values, the comprehensive importance value, the test combination dimension, and the test constraint set to generate a simulation test case set, specifically including: S31, Based on the test combination dimension, pair the scene elements in the scene element structure model in pairs to calculate possible element combinations; After obtaining the test combination dimension and the scene element structure model, the automotive simulation test scene generation device pairs the scene elements in the scene element structure model according to the test combination dimension to calculate possible element combinations. In other words, after obtaining the test combination dimension and scene element structure model, the scene elements in the scene element structure model are paired up in pairs, and all possible combinations are calculated to obtain possible element combinations.

[0041] S32, based on the set of constraints, eliminate invalid combinations from the possible combinations of elements to obtain valid pairwise combinations; After obtaining the set of constraints and the possible combinations of elements, the vehicle simulation test scenario generation device eliminates invalid combinations from the possible combinations of elements based on the set of constraints, and obtains valid pairwise combinations. In other words, after obtaining the set of constraints and the effective pairwise combinations, invalid combinations in the possible element combinations are eliminated based on the constraints in the set of constraints, thus obtaining the total number of pairwise combinations that need to be covered, which is also the effective pairwise combinations.

[0042] S33, mark the valid pairwise combinations as uncovered to obtain a list of uncovered combinations; After obtaining valid pairwise combinations, the automotive simulation test scenario generation device marks the valid pairwise combinations as uncovered, thus obtaining a list of uncovered combinations.

[0043] In other words, after obtaining the set of constraints and the valid pairwise combinations, the list of uncovered combinations is initialized, and the generated valid pairwise combinations are marked as uncovered. The total number of uncovered combinations is... .

[0044] S34, the candidate test cases are filtered according to the uncovered combination list, the scene element structure model, the comprehensive importance value, the element parameter values ​​of the scene elements and the uncovered combination list to obtain a simulation test case set; After acquiring the scene elements and element parameter values, the automotive simulation test scenario generation device filters the candidate test cases according to the uncovered combination list, the scene element structure model, the comprehensive importance value, the element parameter values ​​of the scene elements, and the uncovered combination list to obtain a simulation test case set; wherein, all selected test case sets T are output, and the output format can be JSON structured data format.

[0045] like Figure 5 As shown, the simulation test case set is obtained by filtering the candidate test cases based on the uncovered combination list, the scene element structure model, the comprehensive importance value, the element parameter values ​​of the scene elements, and the uncovered combination list. This includes: generating a test set; comparing the test set with the uncovered combination list; if no pairwise combination in the test set is completely covered by any pairwise combination in the uncovered combination list, then generating all possible candidate test cases under the constraint set through the Cartesian product of the scene element and the element parameter values; obtaining the test case complexity based on the scene element structure model, the element parameter values ​​of the scene elements, and the comprehensive importance value; selecting the candidate test cases with higher test case complexity and covering more uncovered combinations among the element parameter values ​​as the optimal test cases; adding the optimal test cases to the test set and comparing the test set with the uncovered combination list. After comparing the test set with the list of uncovered combinations, the following steps can be taken: if every pair of combinations in the test set is completely covered by every pair of combinations in the list of uncovered combinations, then the test set is used as a set of simulation test cases; if every pair of combinations in the test set is not completely covered by every pair of combinations in the list of uncovered combinations, then a loop is executed to generate all possible candidate test cases under the constraint condition set through the Cartesian product of the scene element and the value of the element parameter.

[0046] As one implementation, comparing the test set with the list of uncovered combinations may include: comparing each pair of combinations in the test set with each pair of combinations in the list of uncovered combinations to determine whether each pair of combinations in the test set is completely covered by each pair of combinations in the list of uncovered combinations.

[0047] As one implementation method, generating all possible candidate test cases under the constraint set through the Cartesian product of the scene elements and the element parameter values ​​can include: generating test cases based on the comprehensive importance of the element parameter values ​​of the scene elements. The test cases can be described as follows: ; In the formula, This represents the overall importance value of the feature parameter.

[0048] As one implementation method, the test case complexity is obtained based on the scene element structure model, the element parameter values ​​of the scene elements, and the comprehensive importance value, and may include: The complexity of test cases can be expressed as: ; In the formula, For the test scenario The complexity of each scenario; For the first The overall importance value of each element in a scenario.

[0049] Selecting candidate test cases with higher complexity and coverage of more uncovered combinations from among the element parameter values ​​as the optimal test case can include: prioritizing candidate test cases with higher complexity and coverage of more uncovered combinations, i.e., the candidate test case with the highest composite priority, and selecting the candidate test case with the highest composite priority as the optimal test case. The specific formula is as follows: ; In the formula, For the first The composite preference value of each candidate test case; For the first The complexity of each test case; The maximum complexity among the candidate test cases; The minimum complexity among the candidate test cases; For the first The number of candidate test cases that cover the number of uncovered combinations; This represents the maximum number of uncovered / uncovered combinations among the candidate test cases; This represents the minimum number of uncovered combinations among the candidate test cases. This is a weighting factor, the value of which can be determined according to the actual situation, and is generally 30%, 50%, 70%, etc. The maximum complexity among candidate test cases. Minimum complexity among candidate test cases Test case complexity can be used to assess the complexity of the test cases. The calculation formula is used to obtain the result.

[0050] This application improves the PICT algorithm to systematically cover all key parameter interactions with a smaller test case set, which can significantly improve testing efficiency.

[0051] S40, Perform scenario simulation on the simulation test case set to generate a simulation test scenario.

[0052] After obtaining the set of simulation test cases, the automotive simulation test scenario generation device performs scenario simulation on the set of simulation test cases to generate simulation test scenarios.

[0053] In other words, to visualize test cases, Calar / Prescan software is used as the simulation software. Simulated test case sets are injected into the simulation software platform through Calar / Prescan's API module, and then visualized to obtain the simulated test scenarios. Taking turning right through an unlit T-junction as an example, some scenario elements of this test case are as follows: Figure 6 As shown, test cases are automatically converted into simulated test scenarios using API scripts. Some scenario snippets are shown below. Figure 7 As shown in the image, the white vehicle is an autonomous vehicle.

[0054] In summary, this method involves obtaining a multi-level scene element structure model composed of several scene elements, as well as acquiring the element parameter values, relative importance of the scene elements at their respective levels, test combination dimensions, and a set of test constraints for each scene element. Based on the relative importance and the scene element structure model, a comprehensive importance weight for each scene element is obtained. A pairwise combination testing algorithm is then used to process the scene element structure model, the scene elements, the element parameter values, the comprehensive importance value, the test combination dimension, and the set of test constraints to generate a set of simulation test cases. Scene simulation is then performed on the set of simulation test cases to generate simulation test scenarios. By deconstructing the connotation of the test scenarios and introducing a comprehensive importance weight for scene elements to improve traditional test case generation algorithms, highly comprehensive and practical test scenarios can be generated quickly and effectively, thus supporting the testing and verification of intelligent connected vehicles.

[0055] This application effectively addresses the problems of high cost and low efficiency in manually building simulation test scenarios, while also resolving the issues of low coverage and insufficient practicality in automatically generated simulation test scenarios using traditional algorithms. This invention has a wide range of applications, capable of deconstructing the scenario's content based on actual testing needs and improving traditional test case generation algorithms by introducing a comprehensive importance assessment of scenario elements. This allows for the rapid and targeted generation of highly comprehensive test scenarios, supporting the testing and verification of intelligent connected vehicles. It helps enterprises accelerate the R&D cycle of intelligent connected vehicles, reduce development costs, and promote the commercialization of autonomous driving technology for intelligent connected vehicles.

[0056] For those consistent with the above, please refer to Figure 8 , Figure 8 This application provides a schematic diagram of the structure of an automotive simulation test scene generation device. (See attached diagram.) Figure 8 As shown, the device includes: The parameter acquisition module 801 is used to acquire a multi-level scene element structure model composed of several scene elements, and to acquire the element parameter values ​​of the scene elements, the relative importance of the scene elements in their respective levels, the test combination dimension, and the set of test constraints. The importance determination module 802 is used to obtain the comprehensive importance weight of scene elements based on the relative importance and the scene element structure model; The test case set determination module 803 is used to process the scene element structure model, the scene elements, the element parameter values, the comprehensive importance value, the test combination dimension, and the test constraint set through a pairwise combination test algorithm to generate a simulation test case set; The scenario simulation module 804 is used to perform scenario simulation on the simulation test case set and generate simulation test scenarios.

[0057] This application also provides a terminal device, which includes: at least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor. When the processor executes the computer program, it implements the steps in the embodiment of the vehicle simulation test scenario generation method.

[0058] This application also provides a computer storage medium storing a computer program for electronic data interchange, which causes a computer to perform some or all of the steps of any of the automotive simulation test scenario generation methods described in the above method embodiments.

[0059] This application also provides a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program that causes a computer to perform some or all of the steps of any of the automotive simulation test scenario generation methods described in the above method embodiments.

[0060] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable storage medium can include at least: any entity or device capable of carrying computer program code to a device / terminal equipment, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. Examples include USB flash drives, portable hard drives, magnetic disks, or optical disks. In some jurisdictions, according to legislation and patent practice, computer-readable storage media cannot be electrical carrier signals or telecommunication signals.

[0061] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0062] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0063] In the embodiments provided in this application, it should be understood that the disclosed devices / terminal equipment and methods can be implemented in other ways. For example, the device / terminal equipment embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling or direct coupling or communication connection may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0064] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

Claims

1. A method for generating automotive simulation test scenarios, characterized in that, include: Obtain a multi-level scene element structure model consisting of several scene elements, and obtain the element parameter values ​​of the scene elements, the relative importance of the scene elements in their respective levels, the test combination dimension, and the set of test constraints. Based on the relative importance and the scene element structure model, the comprehensive importance weight of the scene elements is obtained; The scene element structure model, the scene elements, the element parameter values, the comprehensive importance value, the test combination dimension, and the test constraint set are processed by the pairwise combination test algorithm to generate a set of simulation test cases. The simulation test case set is used to perform scenario simulation to generate simulation test scenarios.

2. The method for generating automotive simulation test scenarios according to claim 1, characterized in that, The process involves using a pairwise combination testing algorithm to process the scene element structure model, the scene elements, the element parameter values, the comprehensive importance value, the test combination dimension, and the set of test constraints to generate a simulation test case set, including: Based on the test combination dimension, the scene elements in the scene element structure model are paired up in pairs to calculate possible element combinations. Based on the set of constraints, invalid combinations are eliminated from the possible combinations of elements to obtain valid pairwise combinations; Mark the valid pairwise combinations as uncovered to obtain a list of uncovered combinations; The simulation test case set is obtained by filtering the candidate test cases based on the uncovered combination list, the scene element structure model, the comprehensive importance value, the element parameter values ​​of the scene elements, and the uncovered combination list.

3. The method for generating automotive simulation test scenarios according to claim 2, characterized in that, The simulation test case set is obtained by filtering the candidate test cases based on the uncovered combination list, the scene element structure model, the comprehensive importance value, the element parameter values ​​of the scene elements, and the uncovered combination list, including: Generate a test suite; Compare the test set with the list of uncovered combinations; If no pairwise combination in the test set is fully covered by all pairwise combinations in the uncovered combination list, then all possible candidate test cases under the constraint condition set are generated by the Cartesian product of the scene element and the element parameter value. The complexity of the test cases is obtained based on the scene element structure model, the element parameter values ​​of the scene elements, and the comprehensive importance value. Select the candidate test cases with higher test case complexity and coverage of more uncovered combinations among the element parameter values ​​as the optimal test cases; Add the optimal test case to the test set, and compare the test set with the list of uncovered combinations; If every pair of combinations in the test set is completely covered by every pair of combinations in the list of uncovered combinations, then the test set will be used as a set of simulation test cases. If no pairwise combination in the test set is completely covered by all pairwise combinations in the uncovered combination list, a loop is executed to generate all possible candidate test cases under the constraint condition set through the Cartesian product of the scene element and the element parameter value.

4. The method for generating automotive simulation test scenarios according to claim 3, characterized in that, Among the selected parameter values, candidate test cases with higher test case complexity and covering more uncovered combinations are chosen as the optimal test cases, including: ； in, For the first The composite preference value of each candidate test case; For the first The complexity of each test case; The maximum complexity among the candidate test cases; The minimum complexity among the candidate test cases; For the first The number of candidate test cases that cover the number of uncovered combinations; This represents the maximum number of uncovered / uncovered combinations among the candidate test cases; This represents the minimum number of uncovered / covered combinations among the candidate test cases. This is the weighting factor.

5. The method for generating automotive simulation test scenarios according to any one of claims 1 to 4, characterized in that, Based on the relative importance and the scene element structure model, the comprehensive importance weight of the scene elements is obtained, including: Based on the relative importance of the scene elements in the scene element structure model at their respective levels, the relative importance matrix information is obtained. Calculate the eigenvector corresponding to the largest eigenvalue of the relative importance matrix at each level in the relative importance matrix information; The relative importance weights are obtained by normalizing the feature vectors. The overall importance weight of scene elements is obtained based on the relative importance weight and the overall importance weight of the previous layer.

6. The method for generating automotive simulation test scenarios according to any one of claims 1 to 4, characterized in that, Before acquiring a multi-level scene element structure model composed of several scene elements, and before acquiring the element parameter values ​​of the scene elements, the relative importance of the scene elements in their respective levels, the test combination dimension, and the set of test constraints, the process includes: Obtain the functional requirements for intelligent connected vehicle simulation testing; Based on the functional requirements and the pre-set functional requirements database, a multi-level scene element structure model consisting of several scene elements is obtained. Display the scene element structure model; Receive the element parameter values ​​of the scene elements returned by the displayed scene element structure model, the relative importance of the scene elements at their respective levels, the test combination dimension, and the set of test constraints.

7. The method for generating automotive simulation test scenarios according to claim 6, characterized in that, Based on the functional requirements and the pre-set functional requirements database, a multi-level scene element structure model consisting of several scene elements is obtained, including: Semantic analysis is performed on the aforementioned functional requirements to obtain requirement scenario information; The required scenario information is matched and analyzed with a preset functional requirement database from two dimensions: traffic environment and test vehicle, to obtain a multi-level scenario element structure model composed of several scenario elements.

8. A device for generating automotive simulation test scenarios, characterized in that, include: The parameter acquisition module is used to acquire a multi-level scene element structure model composed of several scene elements, as well as to acquire the element parameter values ​​of the scene elements, the relative importance of the scene elements in their respective levels, the test combination dimension, and the set of test constraints. The importance determination module is used to obtain the comprehensive importance weight of scene elements based on the relative importance and the scene element structure model; The test case set determination module is used to process the scene element structure model, the scene elements, the element parameter values, the comprehensive importance value, the test combination dimension, and the test constraint set through a pairwise combination test algorithm to generate a simulation test case set. The scenario simulation module is used to perform scenario simulation on the simulation test case set and generate simulation test scenarios.

9. A terminal device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the automotive simulation test scenario generation method as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the method for generating automotive simulation test scenarios as described in any one of claims 1 to 7.

Images