Test method, device, processor and electronic device for vehicle

Abstract

The application discloses a test method, device, processor and electronic equipment of a vehicle, and the method comprises the following steps: in response to detecting a current configuration parameter of a current vehicle, obtaining initial feature data generated in the process of testing a plurality of historical vehicles respectively, and obtaining a plurality of initial feature data; determining target feature data based on the plurality of initial feature data respectively, and obtaining a plurality of target feature data; determining current test data of the current vehicle based on the plurality of target feature data and the current configuration parameter, wherein the current test data is used to represent data required for testing the current vehicle; testing the current vehicle by using the current test data, and obtaining a test result, wherein the test result is used to represent that the function of the current vehicle is a normal function or an abnormal function. The application solves the technical problem that the test of the vehicle has large limitations.

Description

Technical Field

[0001] This application relates to the field of vehicles, and more specifically, to a method, apparatus, processor, and electronic device for testing vehicles. Background Technology

[0002] Currently, vehicle testing often employs traditional onboard test data reuse techniques, reusing historical test data to test vehicles. However, this testing method, by directly reusing and referencing historical test data to test different vehicles, struggles to adapt historical test data to different vehicles, resulting in significant limitations in vehicle testing.

[0003] There is currently no effective solution to the technical problem of the significant limitations in testing the aforementioned vehicles. Summary of the Invention

[0004] This application provides a vehicle testing method, apparatus, processor, and electronic device to at least address the technical problem of significant limitations in vehicle testing.

[0005] According to one aspect of the embodiments of this application, a vehicle testing method is provided. The method includes: in response to detecting current configuration parameters of the current vehicle, acquiring initial feature data generated during the testing of multiple historical vehicles, obtaining multiple initial feature data, wherein the current configuration parameters are used to represent parameters required for functional configuration of the current vehicle, and the initial feature data are used to represent the initial features possessed by the historical vehicle corresponding to the initial feature data under different testing dimensions; determining target feature data based on the multiple initial feature data, obtaining multiple target feature data, wherein the multiple target feature data correspond one-to-one with the multiple initial feature data, and the sensitivity of the target feature data to historical vehicles in terms of data security is lower than the sensitivity of the initial feature data corresponding to the target feature data to historical vehicles in terms of data security; determining current test data of the current vehicle based on the multiple target feature data and the current configuration parameters, wherein the current test data is used to represent the data required for testing the current vehicle; and testing the current vehicle using the current test data to obtain a test result, wherein the test result is used to indicate whether the function of the current vehicle is a normal function or an abnormal function.

[0006] Optionally, target feature data is determined based on multiple initial feature data, including: dividing the initial feature data to obtain first initial feature data and second initial feature data, wherein the first initial feature data has a higher sensitivity to historical vehicles in terms of data security than the second initial feature data; and determining target feature data based on the first initial feature data and the second initial feature data.

[0007] Optionally, the target feature data is determined based on the first initial feature data and the second initial feature data, including: desensitizing the first initial feature data and locking the second initial feature data; merging the desensitized first initial feature data and the locked second initial feature data to obtain the target feature data.

[0008] Optionally, based on multiple target feature data and current configuration parameters, the current test data of the current vehicle is determined, including: generating historical configuration parameters for multiple historical vehicles using multiple target feature data respectively, obtaining multiple historical configuration parameters, wherein the historical configuration parameters are used to represent the parameters required for functional configuration of historical vehicles; determining the difference matrix between the current configuration parameters and multiple historical configuration parameters, and generating a data generation model for determining the current test data based on the multiple historical configuration parameters; and determining the current test data based on the difference matrix, the current configuration parameters, and the data generation model.

[0009] Optionally, a difference matrix is ​​determined between the current configuration parameters and multiple historical configuration parameters, and a data generation model is generated based on the multiple historical configuration parameters to determine the current test data. This includes: comparing the differences between the current configuration parameters and multiple historical configuration parameters to obtain a difference matrix; fusing the multiple historical configuration parameters; deleting invalid configuration parameters from the fused multiple historical configuration parameters; and generating the multiple historical configuration parameters after deleting invalid configuration parameters into a data generation model. In this model, the redundancy of invalid configuration parameters in terms of data redundancy for historical vehicles is higher than the redundancy of the fused historical configuration parameters (excluding invalid configuration parameters) in terms of data redundancy for historical vehicles.

[0010] Optionally, the current test data is determined based on the difference matrix, the current configuration parameters, and the data generation model, including: calibrating the current configuration parameters using the difference matrix; and determining the current test data based on the calibrated current configuration parameters, multiple historical configuration parameters, and the data generation model.

[0011] Optionally, based on the calibrated current configuration parameters, multiple historical configuration parameters, and the data generation model, the current test data is determined, including: inputting the calibrated current configuration parameters and multiple historical configuration parameters into the data generation model; in the data generation model, mapping the multiple historical configuration parameters to the configuration space where the calibrated current configuration parameters are located to obtain the current test data.

[0012] According to one aspect of the embodiments of this application, a vehicle testing apparatus is provided. The apparatus may include: an acquisition unit, configured to, in response to detecting the current configuration parameters of the current vehicle, acquire initial feature data generated during the testing of multiple historical vehicles, thereby obtaining multiple initial feature data, wherein the current configuration parameters represent the parameters required for functional configuration of the current vehicle, and the initial feature data represent the initial features possessed by the historical vehicle corresponding to the initial feature data under different testing dimensions; a first determination unit, configured to, based on the multiple initial feature data, determine target feature data, thereby obtaining multiple target feature data, wherein the multiple target feature data correspond one-to-one with the multiple initial feature data, and the sensitivity of the target feature data to historical vehicles in terms of data security is lower than the sensitivity of the initial feature data corresponding to the target feature data to historical vehicles in terms of data security; a second determination unit, configured to, based on the multiple target feature data and the current configuration parameters, determine the current test data of the current vehicle, wherein the current test data represents the data required for testing the current vehicle; and a testing unit, configured to, using the current test data, test the current vehicle and obtain a test result, wherein the test result indicates whether the function of the current vehicle is a normal function or an abnormal function.

[0013] According to another aspect of the embodiments of this application, a processor is also provided. The processor is used to run a program, wherein the program is executed by the processor to perform the methods described in the embodiments of this application.

[0014] According to another aspect of the embodiments of this application, an electronic device is also provided, including: a memory storing an executable program; and a processor for running the program, wherein the program executes the methods in various embodiments of this application when it runs.

[0015] According to another aspect of the embodiments of this application, a computer-readable storage medium is also provided. This computer-readable storage medium includes a stored program, wherein, when the program is executed, it controls the device where the computer-readable storage medium is located to perform the methods of the embodiments of this application.

[0016] According to another aspect of the embodiments of this application, a computer program product is also provided, the computer program product including a computer program, wherein the computer program implements the method in the embodiments of this application when executed by a processor.

[0017] According to another aspect of the embodiments of this application, a computer program product is also provided, including a non-volatile computer-readable storage medium for storing a computer program, which, when executed by a processor, implements the method in the embodiments of this application.

[0018] According to another aspect of the embodiments of this application, a computer program is also provided, which, when executed by a processor, implements the methods of the embodiments of this application.

[0019] In this embodiment of the application, in response to detecting the current configuration parameters of the current vehicle, initial feature data generated during the testing of various historical vehicles are obtained to obtain multiple initial feature data; target feature data are determined based on the multiple initial feature data to obtain multiple target feature data; current test data of the current vehicle is determined based on the multiple target feature data and the current configuration parameters; and the current test data is used to test the current vehicle to obtain test results. Since this application embodiment, based on obtaining multiple initial feature data, can determine target feature data that is less sensitive to historical vehicles in terms of data security than the initial feature data, that is, obtain multiple target feature data that are adaptive to the current vehicle, and the sensitivity of the aforementioned target feature data to historical vehicles is lower than the sensitivity of historical test data directly reused in related technologies to historical vehicles, by combining the determined multiple target feature data and the detected current configuration parameters, the data required for testing the current vehicle can be determined, and the current vehicle can be tested using the data required for testing the current vehicle. Since the sensitivity of the aforementioned target feature data to historical vehicles has been reduced, and the aforementioned current configuration parameters are the parameters required for configuring the current vehicle's functions, and the current test data determined by combining the two is adapted to the current vehicle, the purpose of generating current test data adapted to the current vehicle is achieved, thereby solving the technical problem of large limitations in vehicle testing, and thus achieving the technical effect of reducing the limitations of vehicle testing. Attached Figure Description

[0020] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0021] Figure 1 This is a flowchart of a vehicle testing method according to an embodiment of this application;

[0022] Figure 2 This is a schematic diagram of a multi-model intelligent driving real-vehicle test data sharing and reuse system based on federated learning and data desensitization, according to an embodiment of this application.

[0023] Figure 3 This is a flowchart of a graded intelligent desensitization method for real vehicle test data according to an embodiment of this application;

[0024] Figure 4 This is a flowchart of a method for safe sharing of multi-vehicle federated learning and adaptive reuse of cross-vehicle data according to an embodiment of this application;

[0025] Figure 5 This is a flowchart of a cross-vehicle adaptive migration and reuse method according to an embodiment of this application;

[0026] Figure 6 This is a schematic diagram of a vehicle testing apparatus according to an embodiment of this application;

[0027] Figure 7 This is a schematic diagram of an electronic device according to an embodiment of this application. Detailed Implementation

[0028] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present application.

[0029] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0030] According to an embodiment of this application, an embodiment of a vehicle testing method is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0031] It should be noted that all information and data involved in this application (including but not limited to initial feature data and current test data) are information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, use and processing of such data must comply with the relevant laws, regulations and standards of the relevant countries and regions, and corresponding operation portals are provided for users to choose to authorize or refuse.

[0032] Figure 1 This is a flowchart of a vehicle testing method according to an embodiment of this application, such as... Figure 1 As shown, the method may include the following steps.

[0033] Step S101: In response to detecting the current configuration parameters of the current vehicle, obtain the initial feature data generated during the testing of various historical vehicles, and obtain various initial feature data.

[0034] In the technical solution provided in step S101 of this application, the aforementioned current configuration parameters can be used to represent the parameters required for configuring the functions of the current vehicle. For example, the aforementioned current configuration parameters can be the full configuration parameters of the current new vehicle model, wherein the aforementioned full configuration parameters can be the following multi-dimensional parameters: sensor parameters, chassis parameters, and intelligent driving parameters, etc., which are only examples here and are not specifically limited.

[0035] In this embodiment, the aforementioned initial feature data can be used to represent the initial features possessed by the historical vehicle under different test dimensions. The aforementioned test dimensions can represent the multiple dimensions of testing the historical vehicle, and the aforementioned initial features can be features under multiple dimensions. Therefore, the aforementioned initial feature data is multi-dimensional real-vehicle test raw data, reflecting the multi-dimensional real-vehicle testing of the historical vehicle and the comprehensiveness of the actual testing. For example, the aforementioned multi-dimensional real-vehicle test raw data may include: perception data, chassis data, intelligent driving decision data, and network connectivity data from real-vehicle road tests, etc. The aforementioned multi-dimensional real-vehicle test raw data can also be simply referred to as real-vehicle test raw data; this is only an example and not a specific limitation.

[0036] In this embodiment, the aforementioned historical vehicles can be used to refer to vehicles that have completed testing within a historical time period. These historical vehicles can be test vehicles mass-produced by any platform that need to complete testing within the historical time period. Any platform can be a production platform for producing the vehicles, such as platform A or platform B. Platform A can also be represented as platform A, and platform B can also be represented as platform B; this is merely an example and not a specific limitation.

[0037] In this embodiment, the term "current vehicle" can refer to a vehicle that is prepared for testing during the current time period. Specifically, the "current vehicle" can be a test vehicle produced by another platform and prepared for testing during the current time period. For example, the other platform could be platform C, which can also be represented as platform C. This is merely an example and not a specific limitation.

[0038] In this embodiment, in response to detecting the current configuration parameters of the current vehicle, initial feature data generated during the testing of various historical vehicles is acquired, resulting in multiple initial feature data sets. Optionally, this embodiment detects whether the multi-vehicle intelligent driving real-vehicle test data sharing and reuse system has received the current configuration parameters. If the multi-vehicle intelligent driving real-vehicle test data sharing and reuse system has received the current configuration parameters, the vehicle model equipment layer can be accessed through the multi-vehicle real-vehicle data acquisition unit to obtain multi-dimensional real-vehicle test raw data, thereby achieving the goal of obtaining multiple initial feature data sets.

[0039] Step S102: Based on various initial feature data, determine the target feature data to obtain various target feature data.

[0040] In the technical solution provided in step S102 of this application, the aforementioned multiple target feature data correspond one-to-one with multiple initial feature data, and the sensitivity of the target feature data to historical vehicles in terms of data security is lower than that of the initial feature data corresponding to the target feature data to historical vehicles in terms of data security. For example, the aforementioned multiple target feature data can be real vehicle test data obtained by classifying and de-identifying the original real vehicle test data.

[0041] In this embodiment, after obtaining multiple initial feature data, target feature data is determined based on each initial feature data, resulting in multiple target feature data. Optionally, this embodiment adjusts the sensitivity of the multiple initial feature data to historical vehicles in terms of data security, thereby obtaining multiple target feature data and achieving the goal of reducing the sensitivity of the initial feature data to historical vehicles in terms of data security.

[0042] Optionally, the sensitivity of various sub-initial feature data to historical vehicles in terms of data security is determined, resulting in multiple sensitivities. The initial feature data includes various sub-initial feature data, which can be used to represent the sub-initial features possessed by the historical vehicle corresponding to the initial feature data under different test dimensions. From these multiple sensitivities, a target sensitivity is determined, wherein the target sensitivity is higher than all other sensitivities among the multiple sensitivities. Adjusting the sub-initial feature data corresponding to the target sensitivity and locking the sub-initial feature data among the multiple sub-initial feature data other than those corresponding to the target sensitivity yields the target feature data.

[0043] Step S103: Based on multiple target feature data and current configuration parameters, determine the current test data of the current vehicle.

[0044] In the technical solution provided in step S103 of this application, the aforementioned current test data can be used to represent the data required to test the current vehicle. This data may include: test conditions, test states, test parameters, number of tests, test order, and test dimensions required to test the current vehicle. For example, the aforementioned current test data may be a dedicated test dataset and test case set adapted to a new vehicle model.

[0045] In this embodiment, after obtaining multiple target feature data, the current test data of the current vehicle is determined based on the multiple target feature data and the current configuration parameters. Optionally, this embodiment utilizes the multiple target feature data to generate historical configuration parameters for each historical vehicle, resulting in multiple historical configuration parameters. These historical configuration parameters can be used to represent the parameters required for configuring the functions of historical vehicles. For example, the historical configuration parameters can be the full configuration parameters of historical vehicles of historical models. Based on the current configuration parameters and multiple historical configuration parameters, the current test data of the current vehicle can be determined, thereby achieving the goal of determining the data required for testing the current vehicle.

[0046] Optionally, based on the current configuration parameters and various historical configuration parameters, the current test data for the current vehicle can be determined. For example, the current configuration parameters can be calibrated, and then, according to the calibrated current configuration parameters, various historical configuration parameters can be converted to obtain the current test data, thereby achieving the goal of determining the data required to test the current vehicle.

[0047] Step S104: Using the current test data, test the current vehicle to obtain the test results.

[0048] In the technical solution provided by step S104 of this application, the test result can be used to indicate whether the current vehicle's function is a normal function or an abnormal function. Specifically, the test result can be a first test result, which can be used to indicate that the current vehicle's overall control function is a normal vehicle control function; or the test result can be a second test result, which can be used to indicate that the current vehicle's overall control function is an abnormal vehicle control function. For example, the aforementioned vehicle control functions may include at least one of the following: driving control functions, auxiliary control functions, entertainment control functions, and intelligent driving control functions. The aforementioned first test result can be used to indicate that the current vehicle's driving control function is a normal driving control function, the current vehicle's auxiliary control function is a normal auxiliary control function, the current vehicle's entertainment control function is a normal entertainment control function, and the current vehicle's intelligent driving control function is a normal intelligent driving control function. The aforementioned second test result can be used to indicate that the current vehicle's driving control function is a normal driving control function, the current vehicle's auxiliary control function is an abnormal auxiliary control function, the current vehicle's entertainment control function is a normal entertainment control function, and the current vehicle's intelligent driving control function is a normal intelligent driving control function. This is only an example and is not specifically limited.

[0049] In this embodiment, after determining the current test data of the current vehicle based on multiple target feature data and current configuration parameters, the current test data is used to test the current vehicle and obtain test results. Optionally, the aforementioned current test data is incorporated into the current vehicle's real-vehicle testing process. That is, edge scenarios, extreme fault samples, and special operating condition data migrated from historical vehicles are reproduced and verified on the current vehicle through real-vehicle road testing or field testing. The aforementioned edge scenarios can represent extreme scenarios encountered by historical vehicles, the aforementioned extreme fault samples can represent faults that occurred in historical vehicles under edge scenarios, and the aforementioned special operating condition data can represent the operating data of historical vehicles under special operating conditions. The current vehicle's status while running the current test data is monitored in real time, and sensor responses, decision logic outputs, and vehicle dynamic behavior are recorded. Based on the recorded data, the test results can be determined, that is, it can be determined whether the current vehicle's function is normal or abnormal.

[0050] Optionally, after the current vehicle has been tested, the newly added test data and new defect samples obtained from the above testing process are again subjected to layered preprocessing, graded desensitization and feature extraction, and uploaded to the federated center to continuously iterate and update the global test feature library, thereby achieving the goal of forming a test data governance closed loop that can be continuously updated and optimized for the vehicle.

[0051] In steps S101 to S104 of this application, in response to detecting the current configuration parameters of the current vehicle, initial feature data generated during the testing of various historical vehicles is obtained, resulting in multiple initial feature data; target feature data is determined based on the multiple initial feature data, resulting in multiple target feature data; current test data of the current vehicle is determined based on the multiple target feature data and the current configuration parameters; and the current test data is used to test the current vehicle to obtain test results. Since this application embodiment, based on obtaining multiple initial feature data, can determine target feature data that is less sensitive to historical vehicles in terms of data security than the initial feature data, that is, obtain multiple target feature data that are adaptive to the current vehicle, and the sensitivity of the aforementioned target feature data to historical vehicles is lower than the sensitivity of historical test data directly reused in related technologies to historical vehicles, by combining the determined multiple target feature data and the detected current configuration parameters, the data required for testing the current vehicle can be determined, and the current vehicle can be tested using the data required for testing the current vehicle. Since the sensitivity of the aforementioned target feature data to historical vehicles has been reduced, and the aforementioned current configuration parameters are the parameters required for configuring the current vehicle's functions, and the current test data determined by combining the two is adapted to the current vehicle, the purpose of generating current test data adapted to the current vehicle is achieved, thereby solving the technical problem of large limitations in vehicle testing, and thus achieving the technical effect of reducing the limitations of vehicle testing.

[0052] The method described in this embodiment will be further described below.

[0053] As an optional embodiment, step S102 involves determining target feature data based on multiple initial feature data, including: dividing the initial feature data to obtain first initial feature data and second initial feature data; and determining target feature data based on the first initial feature data and second initial feature data.

[0054] In this embodiment, the aforementioned multiple initial feature data can be divided into: first initial feature data and second initial feature data. The first initial feature data has a higher sensitivity to historical vehicles in terms of data security than the second initial feature data. Specifically, the first initial feature data can represent the privacy features possessed by the corresponding historical vehicle under different testing dimensions, and the second initial feature data can represent the effective features possessed by the corresponding historical vehicle under different testing dimensions. For example, the first initial feature data can be privacy-related confidential data, and the second initial feature data can be test-effective feature data.

[0055] In this embodiment, after obtaining multiple initial feature data, the initial feature data is divided to obtain first initial feature data and second initial feature data.

[0056] Optionally, this embodiment stratifies multiple sub-initial feature data according to multiple sensitivities to obtain first initial feature data and second initial feature data. That is, the sub-initial feature data corresponding to the target sensitivity is divided into first initial feature data, and the sub-initial feature data other than the sub-initial feature data corresponding to the target sensitivity is divided into second initial feature data.

[0057] For example, after the multi-vehicle real-vehicle data acquisition unit in the data acquisition and layered preprocessing layer collects multi-dimensional raw real-vehicle test data, the data cleaning and denoising unit cleans and denoises the collected data, and the data layering and labeling unit labels the cleaned and denoised data in layers. The data layering and identification unit in the hierarchical intelligent desensitization processing layer identifies the layered data in layers to obtain the layering results, that is, privacy-related confidential data and valid test feature data can be obtained.

[0058] In this embodiment, after dividing the initial feature data to obtain the first initial feature data and the second initial feature data, the target feature data is determined based on the first initial feature data and the second initial feature data.

[0059] Optionally, this embodiment merges or splices the first initial feature data and the second initial feature data to obtain the target feature data. Alternatively, the first initial feature data is blurred, and the blurred first initial feature data and the second initial feature data are merged or spliced ​​to obtain the target feature data. This achieves the goal of reducing the sensitivity of the initial feature data to historical vehicles in terms of data security, thereby improving the accuracy of the target feature data.

[0060] For example, privacy-sensitive data and valid test feature data can be merged or combined to obtain real-vehicle test data for historical vehicles. Alternatively, privacy-sensitive data can be obfuscated, and the obfuscated privacy-sensitive data and valid test feature data can be merged or combined to obtain real-vehicle test data for historical vehicles.

[0061] The steps for determining target feature data based on the first initial feature data and the second initial feature data in this embodiment will be further described below.

[0062] As an optional embodiment, determining target feature data based on first initial feature data and second initial feature data includes: desensitizing the first initial feature data and locking the second initial feature data; merging the desensitized first initial feature data and the locked second initial feature data to obtain the target feature data.

[0063] In this embodiment, the aforementioned desensitization can be an irreversible security desensitization logic that needs to be performed on privacy-sensitive data.

[0064] In this embodiment, the aforementioned locking can be a feature locking protection mechanism that needs to be performed on the valid feature data of the test.

[0065] In this embodiment, after obtaining the first initial feature data and the second initial feature data, the first initial feature data is desensitized, and the second initial feature data is locked. Optionally, this embodiment identifies sensitive content from the first initial feature data, desensitizes the identified sensitive content, and locks the fields of the second initial feature data, that is, prohibits any deletion, modification, blurring, or filtering of the second initial feature data.

[0066] For example, the privacy data desensitization branch performs differentiated desensitization security processing on privacy-related confidential data, while the test feature keep-alive branch performs feature locking protection on test valid feature data. This includes irreversible hash replacement of privacy-related confidential data, clearing the confidential vehicle configuration field of the replaced privacy-related confidential data to zero, blurring the geographic location of the cleared privacy-related confidential data, and removing confidential information. Simultaneously, feature locking protection is applied to test valid feature data, preserving its temporal characteristics and dynamic parameters, and performing unmodified keep-alive processing on the test valid feature data.

[0067] In this embodiment, after desensitizing the first initial feature data and locking the second initial feature data, the desensitized first initial feature data and the locked second initial feature data are merged to obtain the target feature data. Optionally, in this embodiment, the desensitized first initial feature data and the locked second initial feature data are spliced ​​together to obtain real-vehicle test data of historical vehicles. This achieves the goal of reducing the sensitivity of the initial feature data to historical vehicles in terms of data security, thereby realizing the technical effect of improving the accuracy of the target feature data.

[0068] For example, the merging and output unit in the hierarchical intelligent desensitization processing layer merges the two data streams and outputs the merged data. That is, the merging and output unit merges the desensitized privacy-sensitive data and the locked test valid feature data.

[0069] The following section further describes the steps of determining the current test data of the current vehicle based on multiple target feature data and current configuration parameters in this embodiment.

[0070] As an optional embodiment, step S103, determining the current test data of the current vehicle based on multiple target feature data and current configuration parameters, includes: generating historical configuration parameters for multiple historical vehicles using multiple target feature data respectively, obtaining multiple historical configuration parameters; determining the difference matrix between the current configuration parameters and the multiple historical configuration parameters, and generating a data generation model for determining the current test data based on the multiple historical configuration parameters; and determining the current test data based on the difference matrix, the current configuration parameters, and the data generation model.

[0071] In this embodiment, the aforementioned historical configuration parameters can be used to represent the parameters required for configuring the functions of historical vehicles. These historical configuration parameters may include both software and hardware functional parameters required for configuring the functions of historical vehicles. For example, the aforementioned historical configuration parameters may be a set of test feature parameters specific to each historical vehicle model.

[0072] In this embodiment, the aforementioned difference matrix can be a difference mapping matrix between the parameters of the new model and the parameters of historically mature models. For example, the aforementioned difference matrix can also be called a model parameter difference mapping matrix.

[0073] In this embodiment, the data generation model described above can be used to determine the current test data. For example, the data generation model can be a general feature library and a general test model for intelligent driving real-vehicle testing.

[0074] In this embodiment, after obtaining multiple target feature data, historical configuration parameters for various historical vehicles are generated using these target feature data, resulting in multiple historical configuration parameters. Then, a difference matrix between the current configuration parameters and the multiple historical configuration parameters is determined, and a data generation model for determining the current test data is generated based on these multiple historical configuration parameters. Optionally, this embodiment performs feature extraction on the multiple target feature data to obtain multiple historical configuration parameters. Then, the multidimensional differences between the current configuration parameters and the multiple historical configuration parameters are determined, and these multidimensional differences are generated into a vehicle model parameter difference mapping matrix. Finally, the multiple historical configuration parameters are aggregated, and the aggregated configuration parameters are used to generate a general feature library and a general test model for intelligent driving real-vehicle testing.

[0075] For example, the system pre-defines distributed federated nodes for multiple vehicle models. Each vehicle model and each test project group corresponds to an independent local federated node, with each node operating independently and its data not being shared. After each local node completes data preprocessing and hierarchical desensitization, it does not transmit the original test data externally. Instead, it extracts high-dimensional feature vectors from the local lossless test features and iterates the test model locally to generate a test feature parameter set specific to each vehicle model. The multi-dimensional parameter difference comparison unit in the multi-vehicle adaptive migration and reuse layer automatically generates a vehicle parameter difference mapping matrix between the multi-dimensional parameters of new models and historical mature models. The federated center performs weighted fusion and anomaly feature annotation on the multi-dimensional test features of multiple models, aggregating and generating a globally unified general feature library and general test model for intelligent driving real-vehicle testing.

[0076] In this embodiment, after determining the difference matrix and generating the data generation model, the current test data is determined based on the difference matrix, the current configuration parameters, and the data generation model.

[0077] Optionally, this embodiment utilizes the aforementioned difference matrix to calibrate the current configuration parameters. Then, in the data generation model, various historical configuration parameters are transformed according to the calibrated current configuration parameters to obtain the current test data. This achieves the goal of determining the data required for testing the current vehicle, thereby improving the accuracy of the current test data.

[0078] The following section further describes the steps of determining the difference matrix between the current configuration parameters and various historical configuration parameters, and generating a data generation model for determining the current test data based on the various historical configuration parameters in this embodiment.

[0079] As an optional implementation method, determining the difference matrix between the current configuration parameters and multiple historical configuration parameters, and generating a data generation model for determining the current test data based on the multiple historical configuration parameters, includes: comparing the differences between the current configuration parameters and multiple historical configuration parameters to obtain a difference matrix, fusing the multiple historical configuration parameters, deleting invalid configuration parameters from the fused multiple historical configuration parameters, and generating the multiple historical configuration parameters with invalid configuration parameters deleted as a data generation model.

[0080] In this embodiment, the redundancy of the aforementioned invalid configuration parameters in terms of data redundancy for historical vehicles is higher than the redundancy of the fused historical configuration parameters (excluding the invalid configuration parameters) in terms of data redundancy for historical vehicles. The aforementioned invalid configuration parameters can be used to represent redundant features in multi-model, multi-dimensional test features.

[0081] In this embodiment, after considering multiple historical configuration parameters, a difference comparison is performed between the current configuration parameters and the multiple historical configuration parameters to obtain a difference matrix. The multiple historical configuration parameters are then weighted and fused using their respective weights. Next, invalid configuration parameters are removed from the fused multiple historical configuration parameters, and the resulting aggregated configuration parameters are used to generate a data generation model.

[0082] For example, the multi-dimensional parameter difference comparison unit in the multi-model adaptive migration reuse layer automatically compares the multi-dimensional parameter differences between new models and historical mature models, generating a model parameter difference mapping matrix. The federated center performs weighted fusion, redundant feature removal, general feature extraction, and abnormal feature annotation on the multi-model multi-dimensional test features, aggregating and generating a globally unified intelligent driving real vehicle test general feature library and general test model. This achieves the goal of determining the difference matrix and generating the data generation model, thus realizing the technical effect of determining the reliability of the difference matrix and data generation model on which the current test data is based.

[0083] The following section further describes the steps of determining the current test data based on the difference matrix, current configuration parameters, and data generation model in this embodiment.

[0084] As an optional implementation method, the current test data is determined based on the difference matrix, the current configuration parameters, and the data generation model, including: calibrating the current configuration parameters using the difference matrix; and determining the current test data based on the calibrated current configuration parameters, multiple historical configuration parameters, and the data generation model.

[0085] In this embodiment, the above calibration is a cross-vehicle adaptation calibration.

[0086] In this embodiment, after determining the difference matrix and generating the data generation model, the vehicle model parameter difference mapping matrix is ​​used to perform cross-vehicle adaptation calibration on all configuration parameters of the current new vehicle model.

[0087] For example, the test feature adaptive calibration unit in the multi-model adaptive migration reuse layer uses the test feature adaptive normalization algorithm to perform sensor accuracy calibration, dynamic parameter normalization, intelligent driving decision threshold compensation, and scenario condition correction on the global test features issued by the federation one by one, thereby achieving the goal of cross-model adaptation calibration of test features.

[0088] In this embodiment, after calibrating the current configuration parameters using a difference matrix, the current test data is determined based on the calibrated current configuration parameters, multiple historical configuration parameters, and a data generation model. Optionally, in the data generation model, the current test data can be obtained by performing configuration space mapping on multiple historical configuration parameters according to the calibrated current configuration parameters. This achieves the goal of determining the data required to test the current vehicle, thereby improving the accuracy of the current test data.

[0089] The following section further describes the steps of determining the current test data based on the calibrated current configuration parameters, various historical configuration parameters, and the data generation model in this embodiment.

[0090] As an optional implementation method, the current test data is determined based on the calibrated current configuration parameters, multiple historical configuration parameters, and a data generation model, including: inputting the calibrated current configuration parameters and multiple historical configuration parameters into the data generation model; in the data generation model, mapping the multiple historical configuration parameters to the configuration space where the calibrated current configuration parameters are located to obtain the current test data.

[0091] In this embodiment, after inputting the calibrated current configuration parameters and various historical configuration parameters into the data generation model, the various historical configuration parameters are mapped to the configuration space where the calibrated current configuration parameters are located in the data generation model to obtain the current test data.

[0092] Optionally, the calibrated full configuration parameters and the test feature parameter sets specific to each historical vehicle model are input into the general feature library and general test model for intelligent driving real-vehicle testing. In the input model, the test feature parameters specific to each historical vehicle model are mapped to the configuration space where the calibrated full configuration parameters are located, and the current test data can be obtained. This achieves the goal of determining the data required to test the current vehicle, thereby achieving the technical effect of improving the accuracy of the current test data.

[0093] For example, after calibration is completed, mature test scenarios, edge condition samples, fault and abnormal features and test verification logic of historical models are mapped and adapted to the software and hardware system of the new model in the general feature library and general test model of intelligent driving real vehicle test. This can generate a special test dataset and test case set adapted to the new model.

[0094] In this embodiment, based on obtaining multiple initial feature data, target feature data that is more sensitive to historical vehicles in terms of data security than the initial feature data can be determined, resulting in multiple target feature data. By combining the determined multiple target feature data and the detected current configuration parameters, the data required to test the current vehicle can be determined, and the current vehicle can be tested using the data required to test the current vehicle. This achieves the goal of generating current test data that is compatible with the current vehicle, thereby solving the technical problem of large limitations in vehicle testing and achieving the technical effect of reducing the limitations of vehicle testing.

[0095] The technical solutions of the embodiments of this application will be illustrated below with reference to preferred embodiments.

[0096] Currently, vehicle testing often employs traditional onboard test data reuse techniques, reusing historical test data to test vehicles. However, this testing method, by directly reusing and referencing historical test data to test different vehicles, struggles to adapt historical test data to different vehicles, resulting in significant limitations in vehicle testing.

[0097] To address the aforementioned technical problems, this application proposes a vehicle testing method. Based on obtaining multiple initial feature data, target feature data with higher sensitivity to historical vehicles in terms of data security than the initial feature data can be determined, resulting in multiple target feature data. By combining the determined target feature data and the detected current configuration parameters, the data required for testing the current vehicle can be determined, and the current vehicle can be tested using the data required for testing the current vehicle. This achieves the goal of generating current test data adapted to the current vehicle, thereby solving the technical problem of large limitations in vehicle testing and achieving the technical effect of reducing the limitations of vehicle testing.

[0098] In this embodiment, a multi-vehicle intelligent driving real-vehicle test data sharing and reuse system based on federated learning and data anonymization is used to execute a multi-vehicle intelligent driving real-vehicle test data sharing and reuse method based on federated learning and data anonymization. This allows for the adaptive reuse of multiple initial feature data to obtain current test data adapted to the current vehicle. For example, Figure 2 This is a schematic diagram of a multi-vehicle intelligent driving real-vehicle test data sharing and reuse system based on federated learning and data anonymization, according to an embodiment of this application. Figure 2 As shown, the system may include: vehicle model equipment layer 200, data acquisition and hierarchical preprocessing layer 201, hierarchical intelligent desensitization processing layer 202, localized federated learning and sharing layer 203, multi-vehicle model adaptive migration and reuse layer 204, and business application layer 205.

[0099] In this embodiment, the vehicle equipment layer 200 may include: mass-produced test vehicles (covering pure electric, hybrid, and fuel vehicles) from different platforms and configurations of the OEM, various types of sensors, chassis domain, intelligent driving domain, and connected terminal hardware devices, providing the system with multi-dimensional raw data input for real vehicle testing.

[0100] In this embodiment, the data acquisition and layered preprocessing layer 201 can be used to collect multi-dimensional real vehicle test raw data by the multi-model real vehicle data acquisition unit 2011, to clean and denoise the collected data by the data cleaning and denoising unit 2012, and to layer and label the cleaned and denoised data by the data layering and labeling unit 2013.

[0101] In this embodiment, the aforementioned hierarchical intelligent desensitization processing layer 202 can be used by the data hierarchical identification unit 2021 to perform hierarchical identification on the hierarchically labeled data to obtain hierarchical results, that is, privacy-related confidential data and test valid feature data can be obtained. The privacy data desensitization branch unit 2022 performs differentiated desensitization security processing on the privacy-related confidential data, the test feature keep-alive branch unit 2023 performs feature locking protection on the test valid feature data, and the merging and output unit 2024 merges the two data streams and outputs the merged data.

[0102] In this embodiment, the aforementioned localized federated learning sharing layer 203 can be used to complete security feature sharing and model iteration. That is, after the localized federated learning sharing layer 203 completes data preprocessing and hierarchical desensitization at the multi-vehicle local federated nodes 2031, it does not transmit the original test data externally. Instead, the feature extraction unit 2032 performs high-dimensional feature vector extraction and local iteration of the test model on the local lossless test features to generate a test feature parameter set specific to each vehicle model. The federated center aggregation unit 2033 performs weighted fusion, redundant feature removal, general feature extraction, and abnormal feature annotation on the multi-vehicle multi-dimensional test features to aggregate and generate a globally unified intelligent driving real vehicle test general feature library and general test model. The global test model distribution unit 2034 then distributes the optimized model and feature library to all local nodes.

[0103] In this embodiment, the multi-model adaptive migration reuse layer 204 may have a built-in model parameter adaptation database 2041, wherein the model parameter adaptation database 2041 may pre-classify and store the following core difference parameters: all model difference parameters, covering sensor parameters, chassis dynamics parameters, intelligent driving domain control software and hardware version, control response threshold and vehicle operating condition constraint parameters. The aforementioned multi-model adaptive migration and reuse layer 204 can automatically compare the multi-dimensional parameter differences between the new model and historical mature models using the multi-dimensional parameter difference comparison unit 2042, generating a model parameter difference mapping matrix. The test feature adaptive calibration unit 2043 then uses a test feature adaptive normalization algorithm to perform sensor accuracy calibration, dynamic parameter normalization, intelligent driving decision threshold compensation, and scenario condition correction on each of the globally distributed test features, completing the cross-model adaptation calibration of the test features. After the above calibration is completed, the cross-model test data adaptation output unit 2044 accurately maps and adapts the mature test scenarios, edge condition samples, fault anomaly features, and test verification logic of historical models to the software and hardware system of the new model, generating a dedicated test dataset and test case set adapted to the new model. This dedicated test dataset and test case set can be used for real-vehicle testing, fault reproduction, and scenario coverage verification of the new model.

[0104] In this embodiment, the aforementioned business application layer 205 can be used to generate test cases adapted to the new vehicle model by the new vehicle model test case generation unit 2051, supplement the existing test scenarios by the edge scenario supplementation unit 2052, reproduce and verify the faults that may occur in the new vehicle model by the fault reproduction and verification unit 2053, and after the test iteration optimization unit 2054 completes the real vehicle test of the new vehicle model using the adapted migration data, the newly added test scenario data, new defect samples, parameter correction data and new working condition data are uploaded to the federated center after undergoing layered preprocessing, hierarchical desensitization and feature extraction, continuously iteratively updating the global test feature library, and continuously optimizing the cross-platform adaptation accuracy of multiple vehicle models, thereby achieving the purpose of forming a test data governance closed loop that can be continuously updated and self-optimized.

[0105] In this embodiment, by implementing a hierarchical intelligent desensitization method for real vehicle test data, the real vehicle test data can be hierarchically classified and desensitized. For example, Figure 3 This is a flowchart of a graded intelligent desensitization method for real vehicle test data according to an embodiment of this application, such as... Figure 3 As shown, the method may include the following steps.

[0106] Step S301: Receive the original real-vehicle test data of each model.

[0107] After receiving the original real-vehicle test data for each model, step S302 is executed to perform layered identification on the received data.

[0108] After performing hierarchical identification on the received data, steps S303 and S304 are executed to perform irreversible hash replacement on privacy-related confidential data and feature locking protection on test valid feature data.

[0109] After performing an irreversible hash replacement on the privacy-related confidential data, steps S305, S306, and S307 are executed to clear the confidential vehicle configuration field of the replaced privacy-related confidential data, blur the geographical location of the cleared privacy-related confidential data, and remove confidential information.

[0110] In this embodiment, irreversible security desensitization is performed on the privacy-sensitive data branch. This irreversible security desensitization includes: irreversible encrypted replacement of the Vehicle Identification Number (VIN) hash, precise latitude and longitude region blurring offset, zeroing of sensitive vehicle configuration fields, and desensitization and masking of the test terminal number. All desensitization algorithms employ a one-way irreversible processing method, making it impossible for external parties to deduce the original sensitive information from the desensitized data, fully meeting automotive data security regulations and the confidentiality requirements of automakers' trade secrets.

[0111] After feature locking protection is performed on the valid test feature data, steps S308, S309 and S310 are executed to retain the temporal features of the valid test feature data, retain the dynamic parameters of the valid test feature data, and perform unmodified keep-alive processing on the valid test feature data.

[0112] In this embodiment, a feature locking protection operation is performed on the branch of valid test feature data. The feature locking protection operation may include locking the fields of sensor time series data, vehicle dynamic response curve, intelligent driving decision planning logic, fault and abnormal features and scene interaction parameters, prohibiting any deletion, modification, blurring or filtering, thereby ensuring that the test features are intact, time-series aligned and feature complete.

[0113] After removing classified information and keeping the valid test feature data alive without modification, step S311 is executed to merge the two data streams and generate a shareable test dataset.

[0114] After generating a shareable test dataset, step S312 is executed to output the shareable test dataset for federated feature extraction and sharing.

[0115] In this embodiment, by implementing a multi-vehicle federated learning secure sharing and cross-vehicle data adaptive reuse method, the following processes can be integrated: multi-vehicle federated secure sharing, vehicle parameter differentiation adaptation, cross-vehicle data adaptive migration, test data reuse, and closed-loop iteration. For example, Figure 4 This is a flowchart illustrating a method for secure sharing and adaptive data reuse across vehicle models using federated learning, according to an embodiment of this application. Figure 4 As shown, the method may include the following steps.

[0116] Step S401: Perform federal security sharing of localized data for multiple vehicle models.

[0117] In this embodiment, the system pre-defines distributed federated nodes for multiple vehicle models. Each vehicle model and each test project group corresponds to an independent local federated node, and the nodes are independent of each other and their data is not shared. After each local node completes data preprocessing and hierarchical desensitization, it does not transmit the original test data externally. It only performs high-dimensional feature vector extraction and local iteration of the test model on the local lossless test features to generate a test feature parameter set specific to each vehicle model. All local nodes only upload the desensitized feature vectors, model iteration gradient parameters, high-value scenario labels, and fault feature sample parameters to the federated center. The federated center does not store or cache any original vehicle data. It is only responsible for weighted fusion, redundant feature removal, general feature extraction, and abnormal feature annotation of multi-vehicle multi-dimensional test features, aggregating and generating a globally unified intelligent driving real-vehicle test general feature library and general test model. The optimized model and feature library are then distributed back to all local nodes, thereby enabling local retention of original data and global sharing of test experience.

[0118] After the localized data of multiple vehicle models is shared in a federally secure manner, step S402 is executed to adaptively migrate and reuse the data across multiple vehicle models.

[0119] In this embodiment, after the new model testing is initiated, the multi-dimensional parameter differences between the new model and historical mature models are automatically compared to generate a model parameter difference mapping matrix. Using a test feature adaptive normalization algorithm, sensor accuracy calibration, dynamic parameter normalization, intelligent driving decision threshold compensation, and scenario condition correction are performed on each of the globally distributed test features, completing the cross-model adaptation calibration of the test features. After calibration, historical mature test scenarios, edge condition samples, fault anomaly features, and test verification logic are accurately mapped and adapted to the new model's hardware and software system, generating a dedicated test dataset and test case set adapted to the new model.

[0120] After adaptively migrating and reusing data across multiple vehicle models, step S403 is executed to continuously iterate and update the global test feature library, and continuously optimize the cross-platform adaptation accuracy of multiple vehicle models.

[0121] In this embodiment, after conducting real-vehicle testing using the adapted migration data, the newly added test scenario data, new defect samples, parameter correction data, and new operating condition data are uploaded to the federated center after undergoing layered preprocessing, graded desensitization, and feature extraction. The global test feature library is continuously updated iteratively, and the cross-platform adaptation accuracy of multiple vehicle models is continuously optimized. This achieves the goal of forming a test data governance closed loop that can be continuously updated and self-optimized.

[0122] In this embodiment, by executing a cross-vehicle adaptive migration reuse method—that is, a multi-vehicle intelligent driving real-vehicle test data sharing and reuse method based on federated learning and data anonymization—various initial feature data can be adaptively reused to obtain current test data adapted to the current vehicle. For example, Figure 5 This is a flowchart of a cross-vehicle adaptive migration and reuse method according to an embodiment of this application, such as... Figure 5 As shown, the method may include the following steps.

[0123] Step S501: Reuse the initial feature data.

[0124] After reusing the initial feature data, step S502 is executed to perform difference matching between the full configuration parameters of the new model and the configuration parameters of historical mature models to obtain the difference mapping matrix.

[0125] After performing a difference matching between the full configuration parameters of the new model and the configuration parameters of historical mature models to obtain the difference mapping matrix, step S503 is executed to adaptively calibrate the test features to obtain the adaptive parameter package for the new model.

[0126] After adaptively calibrating the test features to obtain the adaptive parameter package for the new model, step S504 is executed to reuse the new model across models.

[0127] After reusing vehicles of the new model across different models, step S505 is executed to perform closed-loop iteration on the current test data.

[0128] For example, an OEM might have gasoline-powered vehicles on platform A, hybrid vehicles on platform B, and a brand-new all-electric intelligent vehicle on platform C. These three platforms have significant differences in sensor configurations, chassis parameters, and intelligent driving algorithm versions. Test data for each platform is independent and isolated. The new platform C vehicle requires full-scale real-vehicle testing from scratch, resulting in high testing costs and slow scenario accumulation. However, by deploying the system described in this application, mature test data from platforms A and B can be securely reused for testing the new platform C vehicle. In other words, the deployed system can generate test data for the new platform C vehicle as follows.

[0129] Step 1: Multi-node data acquisition and hierarchical preprocessing. This system is deployed on mass-production test vehicles on platforms A and B respectively, to collect real-time perception data, chassis data, intelligent driving decision data, and network connectivity data from actual vehicle road tests. After cleaning invalid data, privacy-sensitive fields are uniformly marked as valid test fields.

[0130] Step two, local hierarchical desensitization processing. The local terminals of each platform automatically complete the desensitization operation, performing hash replacement and zeroing processing on VIN codes, geographical locations and vehicle model confidential configurations, while fully preserving radar and camera perception time-series characteristics, vehicle braking and steering response data, and various extreme scenario and fault test characteristics.

[0131] Step 3: Iterative Federation Feature Sharing. Local nodes on platforms A and B extract test feature vectors and model parameters, respectively, and upload them to the federated center. The federated center aggregates high-value test scenarios, edge anomaly samples, and fault features from both platforms to generate and distribute a globally applicable test feature library.

[0132] Step 4: Adaptive migration and reuse for new vehicle models. For new models on the C platform, the system retrieves the vehicle adaptation database, matches the differences in sensor, chassis, and intelligent driving parameters across the three platforms, and completes feature data calibration and adaptation through adaptive algorithms. Edge scenarios and fault samples from platforms A and B, such as rain, backlight, no-signal intersections, and sudden obstacles, are migrated to the C platform vehicle testing process, automatically generating test cases and verification logic adapted to the new vehicle models.

[0133] Step 5: Closed-loop iterative optimization. After the C platform models complete real-vehicle testing, the newly added adaptation data and new defect data are anonymized and uploaded again to update the global feature library and continuously optimize the accuracy of multi-model reuse.

[0134] In this embodiment, based on obtaining multiple initial feature data, target feature data that is more sensitive to historical vehicles in terms of data security than the initial feature data can be determined, resulting in multiple target feature data. By combining the determined multiple target feature data and the detected current configuration parameters, the data required to test the current vehicle can be determined, and the current vehicle can be tested using the data required to test the current vehicle. This achieves the goal of generating current test data that is compatible with the current vehicle, thereby solving the technical problem of large limitations in vehicle testing and achieving the technical effect of reducing the limitations of vehicle testing.

[0135] According to an embodiment of this application, a vehicle testing apparatus is also provided. It should be noted that this vehicle testing apparatus can be used to execute a vehicle testing method as described in the embodiments.

[0136] Figure 6 This is a schematic diagram of a vehicle testing apparatus according to an embodiment of this application. Figure 6 As shown, the vehicle testing device 600 may include: an acquisition unit 601, a first determination unit 602, a second determination unit 603, and a testing unit 604.

[0137] The acquisition unit 601 is used to acquire initial feature data generated during the testing of various historical vehicles in response to the detection of the current configuration parameters of the current vehicle, thereby obtaining various initial feature data. The current configuration parameters are used to represent the parameters required for the functional configuration of the current vehicle, and the initial feature data are used to represent the initial features of the historical vehicle corresponding to the initial feature data under different test dimensions.

[0138] The first determining unit 602 is used to determine target feature data based on multiple initial feature data to obtain multiple target feature data. The multiple target feature data corresponds one-to-one with the multiple initial feature data. The sensitivity of the target feature data to historical vehicles in terms of data security is lower than the sensitivity of the initial feature data corresponding to the target feature data to historical vehicles in terms of data security.

[0139] The second determining unit 603 is used to determine the current test data of the current vehicle based on multiple target feature data and current configuration parameters, wherein the current test data is used to represent the data required to test the current vehicle.

[0140] Test unit 604 is used to test the current vehicle using the current test data and obtain test results, wherein the test results are used to indicate whether the current vehicle's function is normal or abnormal.

[0141] Optionally, the first determining unit 602 may include: a partitioning module, used to partition the initial feature data to obtain first initial feature data and second initial feature data, wherein the first initial feature data has a higher sensitivity to historical vehicles in terms of data security than the second initial feature data; and a first determining module, used to determine target feature data based on the first initial feature data and the second initial feature data.

[0142] Optionally, the first determining module may include: a desensitization submodule, used to desensitize the first initial feature data and lock the second initial feature data; and a merging submodule, used to merge the desensitized first initial feature data and the locked second initial feature data to obtain target feature data.

[0143] Optionally, the second determining unit 603 may include: a first generating module, used to generate historical configuration parameters for various historical vehicles using various target feature data respectively, to obtain various historical configuration parameters, wherein the historical configuration parameters are used to represent the parameters required for functional configuration of historical vehicles; a second generating module, used to determine the difference matrix between the current configuration parameters and the various historical configuration parameters, and to generate a data generation model for determining the current test data based on the various historical configuration parameters; and a second determining module, used to determine the current test data based on the difference matrix, the current configuration parameters and the data generation model.

[0144] Optionally, the second generation module may include: a generation submodule, used to compare the differences between the current configuration parameters and multiple historical configuration parameters to obtain a difference matrix, to fuse multiple historical configuration parameters, to delete invalid configuration parameters from the fused multiple historical configuration parameters, and to generate a data generation model from the multiple historical configuration parameters after deleting invalid configuration parameters. In this model, the redundancy of invalid configuration parameters in terms of data redundancy for historical vehicles is higher than the redundancy of fused historical configuration parameters other than invalid configuration parameters in terms of data redundancy for historical vehicles.

[0145] Optionally, the second determining module may include: a calibration submodule for calibrating the current configuration parameters using a difference matrix; and a determining submodule for determining the current test data based on the calibrated current configuration parameters, multiple historical configuration parameters, and a data generation model.

[0146] Optionally, the determination submodule can determine the current test data based on the calibrated current configuration parameters, multiple historical configuration parameters, and the data generation model by performing the following steps: inputting the calibrated current configuration parameters and multiple historical configuration parameters into the data generation model; in the data generation model, mapping the multiple historical configuration parameters to the configuration space where the calibrated current configuration parameters are located to obtain the current test data.

[0147] In this embodiment, a vehicle testing apparatus is provided. The apparatus may include: an acquisition unit, configured to, in response to detecting the current configuration parameters of the current vehicle, acquire initial feature data generated during testing of multiple historical vehicles, thereby obtaining multiple initial feature data. The current configuration parameters represent the parameters required for functional configuration of the current vehicle, and the initial feature data represent the initial features possessed by the corresponding historical vehicle under different testing dimensions. A first determination unit, configured to, based on the multiple initial feature data, determine target feature data, thereby obtaining multiple target feature data. The multiple target feature data correspond one-to-one with the multiple initial feature data, and the target feature data is secure in terms of data security. The sensitivity of historical vehicles is lower than the sensitivity of the initial feature data corresponding to the target feature data to historical vehicles in terms of data security; the second determining unit is used to determine the current test data of the current vehicle based on multiple target feature data and current configuration parameters, wherein the current test data is used to represent the data required to test the current vehicle; the testing unit is used to test the current vehicle using the current test data and obtain the test results, wherein the test results are used to indicate whether the current vehicle's function is normal or abnormal, thereby achieving the purpose of generating current test data adapted to the current vehicle, thus solving the technical problem of large limitations in vehicle testing, and thus achieving the technical effect of reducing the limitations of vehicle testing.

[0148] According to an embodiment of this application, a processor is also provided for running a program, wherein the program is executed by the processor to perform the methods described in the embodiment.

[0149] According to an embodiment of this application, an electronic device is also provided. Figure 7 This is a schematic diagram of an electronic device according to an embodiment of this application, such as... Figure 7 As shown, the electronic device 700 may include a memory 710 and a processor 720, wherein the memory 710 is used to store an executable program; and the processor 720 is used to run the program stored in the memory 710, wherein the program executes the methods in various embodiments of this application when it runs.

[0150] In this application, "multiple" refers to two or more.

[0151] In this application, unless otherwise expressly defined, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection between two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0152] The terms “first,” “second,” “third,” “fourth,” etc., in this application (if present) are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0153] In this application, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, in this application, the character " / " generally indicates that the preceding and following related objects have an "or" relationship.

[0154] According to another aspect of the embodiments of this application, a computer-readable storage medium is also provided. This computer-readable storage medium includes a stored program, wherein, when the program is executed, it controls the device on which the computer-readable storage medium is located to perform the method described in the embodiments.

[0155] Computer-readable storage media, also known as computer storage media, may include data signals propagated in baseband or as part of a carrier wave, carrying readable program code. These propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable storage media can transmit, propagate, or transfer programs for use by or in conjunction with an instruction execution system, apparatus, or device.

[0156] The program code contained in a computer-readable storage medium may be transmitted using any suitable medium, including but not limited to wireless, wired, optical fiber, radio frequency, or any suitable combination thereof.

[0157] According to an embodiment of this application, a computer program product is also provided, which includes a computer program, wherein the computer program, when executed by a processor, implements the method in the embodiment.

[0158] According to an embodiment of this application, a computer program product is also provided, including a non-volatile computer-readable storage medium for storing a computer program, which, when executed by a processor, implements the method described in the embodiment.

[0159] According to an embodiment of this application, a computer program is also provided, which, when executed by a processor, implements the method described in the embodiment.

[0160] According to an embodiment of this application, a vehicle is also provided, including the electronic equipment described in the embodiment of this application.

[0161] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0162] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be 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 couplings, direct couplings, or communication connections may be through some interfaces; indirect couplings or communication connections between units or modules may be electrical or other forms.

[0163] The units described as separate components may or may not be physically separate. Similarly, the components shown as units may or may not be physical units; they may be located in one place or distributed across multiple units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0164] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0165] 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, the technical solution of this application, in essence, or the part that contributes to related technologies, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard drive, magnetic disk, or optical disk.

[0166] The above are merely preferred embodiments of this application. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A method for testing a vehicle, characterized in that, include: In response to detecting the current configuration parameters of the current vehicle, initial feature data generated during the testing of various historical vehicles are obtained, resulting in various initial feature data. The current configuration parameters are used to represent the parameters required for the functional configuration of the current vehicle, and the initial feature data are used to represent the initial features possessed by the historical vehicle corresponding to the initial feature data under different testing dimensions. Target feature data is determined based on various initial feature data, resulting in various target feature data. Each target feature data corresponds one-to-one with a variety of initial feature data. The sensitivity of the target feature data to the historical vehicle in terms of data security is lower than the sensitivity of the initial feature data corresponding to the target feature data to the historical vehicle in terms of data security. Based on the various target feature data and the current configuration parameters, the current test data of the current vehicle is determined, wherein the current test data is used to represent the data required to test the current vehicle; Using the current test data, the current vehicle is tested to obtain test results, wherein the test results are used to indicate whether the current vehicle's function is normal or abnormal.

2. The method according to claim 1, characterized in that, Target feature data are determined based on various initial feature data, including: The initial feature data is divided to obtain first initial feature data and second initial feature data, wherein the sensitivity of the first initial feature data to the historical vehicle in terms of data security is higher than the sensitivity of the second initial feature data to the historical vehicle in terms of data security. The target feature data is determined based on the first initial feature data and the second initial feature data.

3. The method according to claim 2, characterized in that, Determining the target feature data based on the first initial feature data and the second initial feature data includes: The first initial feature data is desensitized, and the second initial feature data is locked; The first initial feature data after desensitization and the second initial feature data after locking are merged to obtain the target feature data.

4. The method according to claim 1, characterized in that, Based on various target feature data and the current configuration parameters, the current test data of the current vehicle is determined, including: By utilizing various target feature data, various historical configuration parameters for each of the historical vehicles are generated, resulting in various historical configuration parameters, wherein the historical configuration parameters are used to represent the parameters required for functional configuration of the historical vehicles; Determine the difference matrix between the current configuration parameters and various historical configuration parameters, and generate a data generation model for determining the current test data based on the various historical configuration parameters; The current test data is determined based on the difference matrix, the current configuration parameters, and the data generation model.

5. The method according to claim 4, characterized in that, Determine the difference matrix between the current configuration parameters and various historical configuration parameters, and generate a data generation model for determining the current test data based on the various historical configuration parameters, including: The current configuration parameters and various historical configuration parameters are compared to obtain the difference matrix. The various historical configuration parameters are then fused. Invalid configuration parameters are deleted from the fused historical configuration parameters. The data generation model is generated from the various historical configuration parameters after the invalid configuration parameters are deleted. The redundancy of the invalid configuration parameters in terms of data redundancy for the historical vehicles is higher than the redundancy of the fused historical configuration parameters (excluding the invalid configuration parameters) in terms of data redundancy for the historical vehicles.

6. The method according to claim 4, characterized in that, Based on the difference matrix, the current configuration parameters, and the data generation model, the current test data is determined, including: The current configuration parameters are calibrated using the difference matrix. The current test data is determined based on the calibrated current configuration parameters, various historical configuration parameters, and the data generation model.

7. The method according to claim 6, characterized in that, Based on the calibrated current configuration parameters, various historical configuration parameters, and the data generation model, the current test data is determined, including: The calibrated current configuration parameters and various historical configuration parameters are input into the data generation model; In the data generation model, various historical configuration parameters are mapped to the configuration space where the calibrated current configuration parameters are located to obtain the current test data.

8. A vehicle testing device, characterized in that, include: The acquisition unit is configured to, in response to detecting the current configuration parameters of the current vehicle, acquire initial feature data generated during the testing of various historical vehicles, and obtain various initial feature data, wherein the current configuration parameters are used to represent the parameters required for the functional configuration of the current vehicle, and the initial feature data are used to represent the initial features possessed by the historical vehicle corresponding to the initial feature data under different test dimensions. The first determining unit is configured to determine target feature data based on the various initial feature data respectively, thereby obtaining the various target feature data, wherein the various target feature data correspond one-to-one with the various initial feature data, and the sensitivity of the target feature data to the historical vehicle in terms of data security is lower than the sensitivity of the initial feature data corresponding to the target feature data to the historical vehicle in terms of data security. The second determining unit is used to determine the current test data of the current vehicle based on multiple target feature data and the current configuration parameters, wherein the current test data is used to represent the data required to test the current vehicle; The testing unit is used to test the current vehicle using the current test data and obtain test results, wherein the test results are used to indicate whether the current vehicle's function is normal or abnormal.

9. A processor, characterized in that, The processor is used to run a program, wherein the program is executed by the processor to perform the method according to any one of claims 1 to 7.

10. An electronic device, characterized in that, include: Memory, which stores executable programs; A processor for running the program, wherein the program, when running, performs the method according to any one of claims 1 to 7.

Images