Vehicle test scene generalization method and device and storage medium
By constructing a test scenario library and selecting key parameter combinations, the problem of insufficient generalization of vehicle test scenarios was solved, enabling vehicles to effectively respond to risks in diverse scenarios.
Patent Information
- Application Number
- CN202511084064.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-08-04
- Publication Date
- 2025-11-07
AI Technical Summary
Existing technologies do not adequately generalize vehicle testing scenarios, resulting in insufficient vehicle capability to cope with different scenarios.
By generating basic and special test scenarios, a test scenario library is built. Test scenarios are then selected and combined according to test requirements. Key parameters are selected to construct effective generalized scenarios, thereby improving the diversity and effectiveness of test scenarios.
It improves the vehicle's ability to respond to risks in different scenarios and increases the diversity and effectiveness of test scenarios.
Smart Images
Figure CN120909941A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] Embodiments of the present application relate to the technical field of vehicle control, in particular to a generalization method, device and storage medium for vehicle test scenarios. BACKGROUND
[0002] In the design and development process of a vehicle, the vehicle needs to be tested by different scenarios. Testing the vehicle by a single basic scenario may lead to insufficient generalization, limited test scenarios and many invalid generalization scenarios, thereby affecting the ability of the produced vehicle to cope with different scenarios. Therefore, how to increase the diversity and effectiveness of test scenarios and improve the sufficiency of generalization is important for improving the ability of the vehicle to cope with risks in different scenarios. SUMMARY
[0003] Embodiments of the present application provide a generalization method, device and storage medium for vehicle test scenarios, which can be used to improve the ability of the vehicle to cope with risks in different scenarios. The technical solution is as follows:
[0004] In one aspect, the present application provides a generalization method for vehicle test scenarios, which comprises:
[0005] generating a basic test scenario according to a functional specification, the basic test scenario comprising a logical scenario and a functional scenario, the functional specification comprising the dynamics law, design requirements and performance indicators of the vehicle;
[0006] generating a special test scenario related to the basic test scenario based on the parameters of the basic test scenario;
[0007] constructing a test scenario library based on the basic test scenario and the special test scenario, the test scenarios in the test scenario library being classified according to categories;
[0008] in response to receiving a test requirement, screening a plurality of test scenarios from the test scenario library based on the test requirement and combining them to obtain a to-be-generalized scenario;
[0009] selecting a key parameter combination for generalization in the to-be-generalized scenario based on the category of the test scenario contained in the to-be-generalized scenario and the test requirement;
[0010] constructing an effective generalization scenario based on the key parameter combination and the test result corresponding to the key parameter combination.
[0011] In another aspect, a generalization device for vehicle test scenarios is provided, which comprises:
[0012] The first generating module is configured to generate a basic test scene according to a function specification, wherein the basic test scene comprises a logic scene and a function scene, and the function specification comprises a dynamic law, a design requirement and a performance index of a vehicle.
[0013] The second generating module is configured to generate a special test scene related to the basic test scene based on parameters of the basic test scene.
[0014] The first constructing module is configured to construct a test scene library based on the basic test scene and the special test scene, wherein test scenes in the test scene library are classified according to categories.
[0015] The screening module is configured to, in response to receiving a test requirement, screen a plurality of test scenes from the test scene library based on the test requirement and combine the test scenes to obtain a to-be-generalized scene.
[0016] The selecting module is configured to select a key parameter combination for generalization in the to-be-generalized scene based on a category of a test scene contained in the to-be-generalized scene and the test requirement.
[0017] The second constructing module is configured to construct an effective generalization scene based on the key parameter combination and a test result corresponding to the key parameter combination.
[0018] In another aspect, a non-transitory computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, and the computer program is loaded and executed by a processor to implement the vehicle test scene generalization method described above.
[0019] In another aspect, a computer program product is provided, wherein the computer program product comprises computer instructions, and the computer instructions are executed by a processor to implement the steps of the vehicle test scene generalization method described above.
[0020] The technical scheme provided in the application at least brings the following beneficial effects:
[0021] The application generates a basic test scene according to a function specification, and generates a special test scene related to the basic test scene based on parameters of the basic test scene. The basic test scene and the special test scene are used to construct a test scene library, thereby increasing the diversity of test scenes. When a test requirement is received, a plurality of test scenes are screened from the test scene library based on the test requirement and combined, thereby obtaining a to-be-generalized scene. The to-be-generalized scene is selected based on a category of a test scene contained in the to-be-generalized scene and the test requirement, thereby selecting a key parameter combination for generalization in the to-be-generalized scene. An effective generalization scene is constructed based on the key parameter combination and a test result corresponding to the key parameter combination. The diversity and effectiveness of test scenes are increased, and the sufficiency of generalization is improved, thereby improving the ability of a vehicle to cope with risks in different scenes. Attached Figure Description
[0022] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0023] Figure 1 This is a schematic diagram of an implementation environment provided in an embodiment of this application;
[0024] Figure 2 This is a flowchart of a generalization method for a vehicle testing scenario provided in an embodiment of this application;
[0025] Figure 3 This is a schematic diagram of a first boundary curve provided in an embodiment of this application;
[0026] Figure 4 This is a schematic diagram of a second boundary curve provided in an embodiment of this application;
[0027] Figure 5 This is a parameter distribution diagram provided in an embodiment of this application;
[0028] Figure 6 This is a schematic diagram of the structure of a generalization device for a vehicle testing scenario provided in an embodiment of this application. Detailed Implementation
[0029] To make the objectives, technical solutions, and advantages of this application clearer, the embodiments of this application will be described in further detail below with reference to the accompanying drawings.
[0030] This application provides a generalization method for vehicle testing scenarios. Please refer to [the relevant documentation]. Figure 1 The diagram illustrates the implementation environment of the method provided in this embodiment. This implementation environment may include a simulation device 11. The simulation device 11 is used to generate basic test scenarios, generate special test scenarios related to the basic test scenarios, fuse extreme scenarios, and generalize scenarios according to test requirements.
[0031] Based on the above Figure 1 The implementation environment shown in this application provides a method for scenario generalization, such as... Figure 2 As shown, taking the application of this method to a simulation device as an example, the method includes steps 201-206.
[0032] In step 201, the simulation device generates a basic test scene according to a functional specification, the basic test scene including a logic scene and a function scene, the functional specification including dynamics rules, design requirements and performance indicators of the vehicle.
[0033] In a possible implementation, the simulation device generates a plurality of basic test scenes according to a preset functional specification, wherein the basic test scene includes a logic scene and a function scene, and the functional specification includes dynamics rules, design requirements and performance indicators of the vehicle.
[0034] Optionally, the logic scene is used to test driving data of the vehicle and a logic structure of vehicle driving, such as a following scene and a lane changing scene, etc., and the function scene is used to test a specific function of the vehicle, such as an emergency braking scene, etc. Illustratively, the simulation device generates parameters of the basic test scene from historical collected vehicle driving data, wherein the vehicle driving data can be collected by a vehicle-mounted high-precision sensor. Optionally, the vehicle-mounted high-precision sensor includes a laser radar, a camera and a navigation device, etc.
[0035] Illustratively, after the basic test scene is generated, each parameter in the basic test scene needs to be counted and stored. Optionally, the number of generated basic scenes can be set according to requirements and experience, or can be adjusted according to actual conditions.
[0036] In step 202, the simulation device generates a special test scene related to the basic test scene based on parameters of the basic test scene.
[0037] In a possible implementation, after a plurality of basic test scenes are generated, the simulation device generates a special test scene related to the basic test scene based on parameters of the basic test scene, including: adjusting the parameters of the basic test scene to obtain adjusted parameters; and constructing the special test scene related to the basic test scene based on the processed parameters.
[0038] Illustratively, the simulation device adjusts the parameters of the basic test scene in a manner including but not limited to adjusting the parameters of the basic test scene to extreme values. After the processed parameters are obtained, the simulation device constructs the special test scene related to the basic test scene based on the processed parameters.
[0039] In step 203, the simulation device constructs a test scene library based on the basic test scene and the special test scene, and test scenes in the test scene library are classified according to categories.
[0040] Optionally, after the generation of the basic test scene and the special test scene is completed, the simulation device constructs a test scene library based on the basic test scene and the special test scene, wherein test scenes in the test scene library are classified according to categories.
[0041] Exemplarily, after the generation of the basic test scenarios and the special test scenarios is completed, the simulation device performs parameter fusion on the basic test scenarios and the special test scenarios to obtain test scenarios; each test scenario is classified and numbered, and the number and test result of each test scenario contained in each classification are recorded and stored; and a test scenario library is constructed based on the classified and numbered test scenarios. In a possible implementation manner, the test scenario library contains category descriptions of various classifications and numbers contained in each classification, where the category description corresponding to each classification can be preset.
[0042] In a possible implementation manner, after the test scenario library is constructed, the simulation device constructs extreme scenarios based on the test scenarios; and the extreme scenarios are fused into the test scenarios. Optionally, the simulation device constructs the extreme scenarios based on the test scenarios, including: constructing the extreme scenarios by introducing at least one of extreme weather or sensor deception, where the manner of sensor deception includes interfering with the recognition result of a sensor by a preset false signal.
[0043] Optionally, the extreme scenarios are constructed by introducing the extreme weather, including: selecting parameters of a basic test scenario with normal weather conditions, adjusting the weather state to a preset extreme weather, and constructing an extreme scenario based on the parameters of the basic test scenario after the weather state adjustment. Optionally, the extreme weather includes but is not limited to: rainfall greater than a rainfall threshold, visibility lower than a visibility threshold, light intensity greater than a light intensity threshold, and wind speed greater than a wind speed threshold. Wherein, the rainfall threshold, the visibility threshold, the light intensity threshold and the wind speed threshold can be preset according to experiments.
[0044] In a possible implementation manner, the extreme scenarios are constructed by the manner of sensor deception, including: determining parameters in the parameters of the basic test scenario that are acquired by a sensor, and interfering with the sensor acquisition process by a pre-designed corresponding interference manner. Exemplarily, the interference manner includes but is not limited to: projecting an adversarial image, for example, a false traffic sign, to a visual acquisition sensor; transmitting an interference signal to a radar; or adding reflective material around a light sensing material, etc.
[0045] In step 204, in response to receiving the test requirement, the simulation device screens multiple test scenarios from the test scenario library based on the test requirement and combines the multiple test scenarios to obtain a to-be-generalized scenario.
[0046] Optionally, after the construction of the test scenario library is completed, if a test requirement is received, the simulation device screens multiple test scenarios from the test scenario library based on the test requirement and combines them to obtain a to-be-generalized scenario. In a possible implementation, a user can input a test requirement through the simulation device, where the test requirement includes but is not limited to a functional requirement of a vehicle, for example, whether to effectively trigger AEB (Autonomous Emergency Braking); a performance requirement of the vehicle, for example, optimization of a braking distance of the vehicle; and a safety requirement of the vehicle, for example, adjustment speed of a driving mode.
[0047] For example, after the test requirement is received, the simulation device screens multiple test scenarios from the test scenario library based on the test requirement and combines them to obtain a to-be-generalized scenario. Optionally, the number of test scenarios used for combination can be set according to experience, for example, can be set to two-by-two combination between test scenarios to obtain multiple to-be-generalized scenarios.
[0048] In step 205, the simulation device selects a key parameter combination used for generalization in the to-be-generalized scenario based on a category of a test scenario contained in the to-be-generalized scenario and the test requirement.
[0049] In a possible implementation, after the to-be-generalized scenario is obtained, the simulation device selects a key parameter combination used for generalization in the to-be-generalized scenario based on a category of a test scenario contained in the to-be-generalized scenario and the test requirement, including: the simulation device determines the key parameter combination used for generalization in the to-be-generalized scenario based on the category of the test scenario contained in the to-be-generalized scenario and the test requirement according to a correspondence between the category of the test scenario and the test requirement and the key parameter combination used for generalization. The key parameter combination usually includes a pair of key parameters.
[0050] Optionally, the correspondence between the category of the test scenario and the test requirement and the key parameter combination used for generalization can be set according to experiments in advance, for example, if the category of the test scenario is a following scenario and the test requirement is a functional requirement of whether to effectively trigger AEB, the key parameter combination used for generalization is a relative speed and a relative distance between vehicles.
[0051] In step 206, the simulation device constructs an effective generalization scenario based on the key parameter combination and a test result corresponding to the key parameter combination.
[0052] For example, after the key parameter combination used for generalization in the to-be-generalized scenario is determined, the simulation device constructs an effective generalization scenario based on the key parameter combination and a test result corresponding to the key parameter combination, including: drawing a parameter distribution diagram of the to-be-generalized scenario with the key parameter combination as coordinates; and constructing the effective generalization scenario based on a distribution of the key parameter combination of different test results in the parameter distribution diagram.
[0053] In a possible implementation, taking a pair of key parameters as an example, the simulation device draws a parameter distribution diagram of the to-be-generalized scene with the pair of key parameters as coordinates, including: the simulation device respectively establishes a coordinate system with the two key parameters as the horizontal and vertical coordinates, and draws a point corresponding to the pair of key parameters in the established coordinate system, and different test results are distinguished by points of different colors, to obtain the parameter distribution diagram of the to-be-generalized scene. Optionally, the drawing tool can be excel (Microsoft spreadsheet), matlab (matrix laboratory), or python.
[0054] Optionally, after the parameter distribution diagram of the to-be-generalized scene is drawn, the simulation device constructs an effective generalization scene based on the distribution of the key parameter combinations of different test results in the parameter distribution diagram, including: determining the boundary points of the key parameter combinations of different test results in the parameter distribution diagram; extracting the key parameter combinations corresponding to the boundary points and performing curve fitting to obtain a first boundary line and a second boundary line; and selecting the key parameter combinations corresponding to the region between the first boundary line and the second boundary line to construct the effective generalization scene.
[0055] Exemplarily, the simulation device determines the boundary points of the key parameter combinations of different test results in the parameter distribution diagram according to the distribution of points of different colors in the parameter distribution diagram. In a possible implementation, after the key parameter combinations corresponding to the boundary points are extracted, the key parameter combinations corresponding to the boundary points are curve fitted, including: the key parameter combinations corresponding to the boundary points are curve fitted by a quintic polynomial, and optionally, the quintic polynomial is as follows:
[0056] ΔV=p1S 5 +p2S 4 +p3S 3 +p4S 2 +p5S+p6
[0057] ΔV and S are the key parameter combinations corresponding to the boundary points, and p1, p2, p3, p4, p5, and p6 are the coefficients of the fitted curve.
[0058] Optionally, after the first boundary line and the second boundary line are fitted, the key parameter combinations corresponding to the region between the first boundary line and the second boundary line are selected to construct the effective generalization scene, including: the first boundary line and the second boundary line are drawn in the parameter distribution diagram, the key parameter combinations contained in the region between the first boundary line and the second boundary line are selected as effective key parameter combinations, and the effective generalization scene is constructed according to the effective key parameter combinations.
[0059] In a possible implementation, the category of the test scenario is a following scenario, the test requirement is a functional requirement of whether to effectively trigger AEB, and the key parameter combination for generalization is taken as an example, the relative speed and the relative distance between vehicles, and the test result includes a collision and no collision.
[0060] Exemplarily, the simulation device establishes a coordinate system with the relative speed as the horizontal coordinate and the relative distance as the vertical coordinate, and depicts the key parameter combination of the test result as no collision with a green dot and the key parameter combination of the test result as a collision with a red dot.
[0061] Optionally, the relative speed of the vehicle is brought into ΔV in the quintic polynomial, the relative distance of the vehicle is brought into S in the quintic polynomial, p1, p2, p3, p4, p5 and p6 are calculated, and then the first boundary line and the second boundary line are fitted. The schematic diagram of the first boundary curve is as shown in Figure 3 The schematic diagram of the second boundary curve is as shown in Figure 4 The parameter distribution diagram is as shown in Figure 5 The area between the first boundary line and the second boundary line is the area corresponding to the effective generalization scenario, and therefore the key parameter combination corresponding to the area between the first boundary line and the second boundary line is selected to construct the effective generalization scenario.
[0062] In a possible implementation, after the construction of the effective generalization scenario is completed, the simulation device eliminates the key parameter combinations other than the key parameter combination corresponding to the effective generalization scenario, and performs simulation testing on the vehicle through the effective generalization scenario. Exemplarily, after the vehicle completes the simulation testing and enters the real vehicle testing stage, the logic problems, the after-sales problems and the driving parameters when the problems occur in the vehicle control process are recorded and stored, and the corresponding basic test scenario is generated based on the driving parameters when the problems occur in the subsequent simulation testing.
[0063] The embodiment of the application generates a basic test scenario through a functional specification, generates a special test scenario related to the basic test scenario based on the parameters of the basic test scenario, constructs a test scenario library through the basic test scenario and the special test scenario, and increases the diversity of the test scenario. When a test requirement is received, multiple test scenarios are selected from the test scenario library based on the test requirement and combined to obtain a to-be-generalized scenario. Then, the key parameter combination for generalization is selected from the to-be-generalized scenario according to the category of the test scenario contained in the to-be-generalized scenario and the test requirement. Finally, the effective generalization scenario is constructed based on the key parameter combination and the test result corresponding to the key parameter combination. The diversity and effectiveness of the test scenario are increased, and the sufficiency of generalization is improved, thereby improving the ability of the vehicle to cope with risks in different scenarios.
[0064] Referring to Figure 6The embodiments provide a generalization device of a vehicle test scene, and the device comprises:
[0065] A first generation module 601 is configured to generate a basic test scene according to a function specification, the basic test scene comprising a logic scene and a function scene, and the function specification comprising dynamics rules, design requirements and performance indexes of a vehicle;
[0066] A second generation module 602 is configured to generate a special test scene related to the basic test scene based on parameters of the basic test scene.
[0067] A first construction module 603 is configured to construct a test scene library based on the basic test scene and the special test scene, and test scenes in the test scene library are classified according to categories.
[0068] A screening module 604 is configured to, in response to receiving a test requirement, screen a plurality of test scenes from the test scene library based on the test requirement and combine the test scenes to obtain a to-be-generalized scene.
[0069] A selection module 605 is configured to select a key parameter combination for generalization in the to-be-generalized scene based on a category of a test scene contained in the to-be-generalized scene and the test requirement.
[0070] A second construction module 606 is configured to construct an effective generalization scene based on the key parameter combination and a test result corresponding to the key parameter combination.
[0071] In a possible implementation, the second construction module 606 is configured to draw a parameter distribution diagram of the to-be-generalized scene with the key parameter combination as a coordinate, and construct the effective generalization scene based on a distribution of the key parameter combination of different test results in the parameter distribution diagram.
[0072] In a possible implementation, the second construction module 606 is configured to determine a boundary point of the key parameter combination of different test results in the parameter distribution diagram, extract the key parameter combination corresponding to the boundary point and perform curve fitting to obtain a first boundary line and a second boundary line, and select the key parameter combination corresponding to a region between the first boundary line and the second boundary line to construct the effective generalization scene.
[0073] In a possible implementation, the second generation module 602 is configured to adjust the parameters of the basic test scene to obtain processed parameters, and construct the special test scene related to the basic test scene based on the processed parameters.
[0074] In a possible implementation, the first construction module 603 is further configured to construct an extreme scene based on the test scene, and fuse the extreme scene into the test scene.
[0075] In a possible implementation, the first construction module 603 is configured to construct the extreme scene by introducing at least one of extreme weather or sensor deception, and the sensor deception includes interfering with a recognition result of the sensor by a preset false signal.
[0076] In a possible implementation, the apparatus further includes an emulation test module configured to perform an emulation test on the vehicle by using the effective generalization scene.
[0077] The apparatus generates a basic test scene according to a function specification, generates a special test scene related to the basic test scene based on parameters of the basic test scene, constructs a test scene library by using the basic test scene and the special test scene, increases the diversity of the test scene, selects multiple test scenes from the test scene library based on a test requirement and combines the multiple test scenes to obtain a to-be-generalized scene, selects a key parameter combination for generalization from the to-be-generalized scene according to a category of a test scene included in the to-be-generalized scene and the test requirement, and constructs an effective generalization scene based on the key parameter combination and a test result corresponding to the key parameter combination. The apparatus increases the diversity and effectiveness of the test scene, improves the sufficiency of generalization, and thus improves the capability of the vehicle to cope with risks in different scenes.
[0078] It should be noted that, when the apparatus provided in the above embodiments implements its functions, only the division of the above functional modules is exemplified, and in actual application, the above functions can be completed by different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules to complete all or part of the above functions. In addition, the apparatus and the method provided in the above embodiments belong to the same concept, and the specific implementation process is described in the method embodiments, which will not be repeated here.
[0079] In an example embodiment, a computer readable storage medium is also provided, and the computer readable storage medium stores at least one computer program. The at least one computer program is loaded and executed by a processor of a computer device, so that the computer implements any one of the above vehicle test scene generalization methods.
[0080] In a possible implementation, the above computer readable storage medium can be a read-only memory (ROM), a random access memory (RAM), a compact disc read-only memory (CD-ROM), a magnetic tape, a floppy disk, and an optical data storage device, etc.
[0081] In an example embodiment, a computer program product or computer program is also provided, which includes computer instructions stored in a computer readable storage medium. A processor of a computer device reads the computer instructions from the computer readable storage medium, and the processor executes the computer instructions, so that the computer device performs the generalization method of any one of the vehicle test scenarios described above.
[0082] It should be noted that the information (including but not limited to user device information, user personal information, etc.), data (including but not limited to data for analysis, stored data, displayed data, etc.) and signals involved in the present application are all authorized by the user or fully authorized by all parties, and the collection, use and processing of related data need to comply with relevant laws, regulations and standards of relevant countries and regions. For example, the basic test scenarios, special test scenarios, test scenario library, test requirements and effective generalization scenarios involved in the present application are obtained under full authorization.
[0083] It should be understood that "multiple" referred to in the present application refers to two or more. "And / or", which describes the association relationship of associated objects, means that there can be three relationships, for example, A and / or B can mean that there are three cases of A alone, A and B together, and B alone. The character " / " generally represents that the associated objects before and after it are in an "or" relationship.
[0084] It should be noted that the terms "first", "second", etc. (if any) in the specification and claims of the present application are used to distinguish similar objects, and do not necessarily describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. The embodiments described in the following example embodiments do not represent all embodiments consistent with the present application. Rather, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.
[0085] The above is only an example embodiment of the present application and does not limit the present application. Any modification, equivalent replacement, improvement, etc. made within the principles of the present application shall be included in the protection scope of the present application.
Claims
1. A method of generalization of vehicle test scenarios, characterized in that, The method comprises: generating a basic test scene according to a functional specification, the basic test scene comprising a logic scene and a functional scene, the functional specification comprising dynamics laws, design requirements and performance indicators of a vehicle; generating a special test scene related to the basic test scene based on parameters of the basic test scene; constructing a test scene library based on the basic test scene and the special test scene, test scenes in the test scene library being classified according to categories; in response to receiving a test requirement, screening and combining a plurality of test scenes from the test scene library based on the test requirement to obtain a to-be-generalized scene; selecting a key parameter combination for generalization in the to-be-generalized scene based on categories of the test scenes contained in the to-be-generalized scene and the test requirement; constructing an effective generalization scene based on the key parameter combination and a test result corresponding to the key parameter combination.
2. The method of claim 1, wherein, The method of constructing an effective generalization scene based on the key parameter combination and a test result corresponding to the key parameter combination comprises: drawing a parameter distribution diagram of the to-be-generalized scene with the key parameter combination as coordinates; constructing the effective generalization scene based on distribution of key parameter combinations of different test results in the parameter distribution diagram.
3. The method of claim 2, wherein, The method of constructing the effective generalization scene based on distribution of key parameter combinations of different test results in the parameter distribution diagram comprises: determining boundary points of key parameter combinations of different test results in the parameter distribution diagram; extracting key parameter combinations corresponding to the boundary points and performing curve fitting to obtain a first boundary line and a second boundary line; selecting key parameter combinations corresponding to a region between the first boundary line and the second boundary line to construct the effective generalization scene.
4. The method of claim 1, wherein, The method of generating a special test scene related to the basic test scene based on parameters of the basic test scene comprises: adjusting parameters of the basic test scene to obtain adjusted parameters; constructing a special test scene related to the basic test scene based on the adjusted parameters.
5. The method of claim 1, wherein, After the test scene library is constructed based on the basic test scene and the special test scene, the method further comprises: constructing an extreme scene based on the test scene; fusing the extreme scene into the test scene.
6. The method of claim 5, wherein, The method of constructing an extreme scene based on the test scene comprises: constructing an extreme scene by introducing at least one of extreme weather or a sensor deception mode, the sensor deception mode comprising interfering with a recognition result of a sensor by a preset false signal.
7. The method of claim 1, wherein, The method further comprises: performing simulation testing on the vehicle by using the effective generalization scene.
8. A generalization apparatus of a vehicle test scenario, characterized by, The device comprises: a first generation module configured to generate a basic test scene according to a functional specification, the basic test scene comprising a logic scene and a functional scene, the functional specification comprising dynamics laws, design requirements and performance indicators of a vehicle; a second generation module configured to generate a special test scene related to the basic test scene based on parameters of the basic test scene; a first construction module configured to construct a test scene library based on the basic test scene and the special test scene, test scenes in the test scene library being classified according to categories; The screening module is configured to, in response to receiving a test requirement, screen a plurality of the test scenarios from the test scenario library based on the test requirement and combine the test scenarios to obtain a to-be-generalized scenario; The selection module is configured to select a key parameter combination for generalization in the to-be-generalized scenario based on a category of a test scenario contained in the to-be-generalized scenario and the test requirement. The second construction module is configured to construct an effective generalization scenario based on the key parameter combination and a test result corresponding to the key parameter combination.
9. A computer program product, comprising computer instructions, which, when executed by a processor, implement the steps of the scenario generalization method according to any one of claims 1 to 6.
10. A non-transitory computer-readable storage medium, comprising: The computer program is stored in the computer readable storage medium and loaded and executed by the processor to implement the scenario generalization method according to any one of claims 1 to 6.