Vehicle test management method, system and device, electronic equipment and storage medium

Abstract

The application relates to a vehicle test management and control method, system and device, electronic equipment and a storage medium. The method comprises the following steps: after the system is triggered to complete initialization, a test task is sent to an edge end, is issued to a terminal by the edge end, so that the terminal receives the test task, calls a physical test cluster and a virtual multi-agent cluster to cooperatively execute the test task, and uploads test full-link data to a cloud end in real time through the edge end during test execution; the cloud end receives the full-link data, performs abnormality identification and risk warning level judgment based on the full-link data; if the risk warning level reaches a triggering intervention action level, a corresponding standardized intervention strategy is matched according to the risk warning level, the terminal receives an intervention instruction, completes a corresponding disposal action, and triggers full-link fault recording; and the cloud end checks the intervention result. The application can realize full-dimension abnormality real-time identification, cloud-end grading real-time warning and automatic closed-loop intervention full-process intelligent management and control.

Description

Technical Field

[0001] This application relates to the field of intelligent connected vehicle and autonomous driving testing technology, and in particular to a vehicle testing management method, system, device, electronic device and storage medium. Background Technology

[0002] With the rapid development of the intelligent connected vehicle industry, the functional complexity and integration of L4 (highly automated driving, a level in the classification of driving automation) and above intelligent connected vehicles have increased significantly, and the demand for full-chain reliability testing is becoming increasingly urgent. In particular, for long-cycle unattended testing, extremely high requirements are placed on the test system's ability to identify anomalies, provide real-time warnings, and perform automated interventions.

[0003] Existing technologies have proposed vehicle-road-cloud integrated driving or traffic monitoring modes. However, in most cases, the vehicle-road-cloud integrated technology mode is mainly used in the following two scenarios: one is to focus on the allocation of vehicle-cloud computing power and optimization of behavior sequence for autonomous driving decision-making and planning under normal driving conditions. It designs a cloud-based collaborative decision-making mechanism only for the scenario of normal driving of connected vehicles and is used for autonomous driving planning under normal driving conditions. The other is a vehicle-road-cloud collaborative, multi-source fusion traffic perception and emergency early warning solution for traffic operation and management scenarios. It realizes the deep cloud fusion of roadside and vehicle-side perception data, accurate identification of traffic abnormal events and hierarchical emergency early warning. This mode is mainly used for perception and early warning of traffic management business.

[0004] Therefore, existing vehicle-road-cloud integrated driving technologies cannot adapt to the vehicle-road-cloud distributed architecture, cover the entire process of autonomous driving testing, and cannot meet the long-cycle, high-reliability testing requirements of L4 and above intelligent connected vehicles. Summary of the Invention

[0005] In order to solve the above-mentioned technical problems, or at least partially solve the above-mentioned technical problems, this application provides a vehicle testing and control method, system, device, electronic device and storage medium.

[0006] In a first aspect, this application provides a vehicle testing method, the method being applied in a cloud environment, the method comprising: After the trigger system completes initialization, the cloud sends test tasks to the edge terminal, which then distributes them to the terminal. Upon receiving the test task, the terminal invokes the physical test cluster and the virtual multi-agent cluster to collaboratively execute the test task. During the test execution, the entire test chain data is uploaded to the cloud in real time via the edge terminal. The edge terminal preprocesses the data before uploading it to the cloud in real time, while simultaneously performing local backups. Receive the end-to-end data and perform anomaly identification and risk warning level determination based on the end-to-end data; If the risk warning level reaches the level that triggers an intervention action, a corresponding standardized intervention strategy is matched according to the risk warning level, and the execution instruction of the standardized intervention strategy is sent to the terminal; so that the terminal receives the instruction, completes the corresponding handling action, and triggers full-link fault recording. The system tracks the execution status of the standardized intervention strategy and verifies the intervention results.

[0007] In one possible embodiment of this application, the tracking system further verifies the execution status of the standardized intervention strategy, including: The cloud acquires execution status data in real time during the execution of the intervention strategy by the terminal, and performs result verification based on the status data to determine whether the anomaly has been eliminated; If so, resume the testing process; If not, continue with end-to-end anomaly identification and risk level assessment until the anomaly is eliminated and the testing process resumes.

[0008] In one possible embodiment of this application, before the triggering system completes the test environment initialization, the following is included: In response to the tester's test task creation operation, the test requirements are parsed and the test case set is obtained; Based on the test case set, the binding work between the terminal and the corresponding edge terminal is performed, the parameterized modeling of the virtual scene and the configuration of test rules are completed, and the configuration information is sent to the terminal and the edge terminal after compliance verification, so that the terminal can complete the spatiotemporal alignment of the physical test environment and the virtual simulation scene, and the edge terminal can complete the timing synchronization and adaptation of the virtual and real scenes.

[0009] In one possible embodiment of this application, the method further includes: If the risk warning level does not reach the level that triggers intervention, then information recording is performed, or alarm information is pushed to the terminal and edge terminal, and the scene recording when the anomaly occurs is saved.

[0010] In one possible embodiment of this application, receiving the end-to-end data and performing anomaly identification and risk warning level determination based on the end-to-end data includes: The system receives end-to-end data from the receiving terminal, covering the perception layer, decision-making layer, planning layer, control layer, multi-agent collaboration layer, and testing process layer. It then uses a composite mechanism of threshold judgment, trend prediction, and AI anomaly recognition to identify anomalies and determine risk levels.

[0011] In one possible embodiment of this application, the method further includes: After the testing is completed, the cloud platform generates standardized test reports and anomaly analysis reports based on the full-process test data and pushes them to the cloud management backend.

[0012] In one possible embodiment of this application, the method further includes: After the testing was completed, the cloud platform used the full-process test data to iteratively optimize the virtual scene library and supplement it with complex dynamic scene samples. Based on the newly added abnormal samples in this test, the AI ​​anomaly recognition model will be incrementally trained and its accuracy evaluated, and the optimized model will be released.

[0013] In one possible embodiment of this application, the method further includes: If the anomaly is not eliminated, further determine whether the current risk level is the highest warning level; If not, then upgrade to the highest risk warning level; If so, the task of executing the highest level of security policy will be triggered.

[0014] Secondly, this application provides a vehicle testing system, the system comprising: a cloud, an edge, and a terminal; The cloud is used to send test tasks to the edge terminal after the trigger system completes initialization, and then distribute the tasks to the terminal via the edge terminal. The terminal is used to receive the test task, call the physical test cluster and the virtual multi-agent cluster to jointly execute the test work, and send the test full-link data to the edge terminal in real time during the test execution process; The edge device preprocesses and backs up the data locally, then uploads it to the cloud in real time. The cloud platform is also used to receive the end-to-end data and to perform anomaly identification and risk warning level judgment based on the end-to-end data. If the risk warning level reaches the level that triggers an intervention action, a corresponding standardized intervention strategy is matched according to the risk warning level, and the execution instruction of the standardized intervention strategy is sent to the terminal; so that the terminal receives the instruction, completes the corresponding handling action, and triggers full-link fault recording. The system tracks the execution status of the standardized intervention strategy and verifies the intervention results.

[0015] Thirdly, this application provides a vehicle testing apparatus, comprising: The test task sending module is used to send test tasks from the cloud to the edge terminal after the trigger system completes initialization, and then distribute the test tasks to the terminal via the edge terminal. After receiving the test task, the terminal calls the physical test cluster and the virtual multi-agent cluster to collaboratively execute the test task, and uploads the test end-to-end data to the cloud in real time via the edge terminal during the test execution process. The edge terminal preprocesses and backs up the data locally before uploading it to the cloud in real time. Anomaly detection module is used to receive the end-to-end data and perform anomaly detection and risk warning level judgment based on the end-to-end data; The intervention strategy matching module is used to match a standardized intervention strategy according to the risk warning level if the risk warning level reaches the level that triggers an intervention action, and send the execution instruction of the standardized intervention strategy to the terminal; so that the terminal can complete the corresponding handling action after receiving the instruction and trigger the full-link fault recording. The verification module is used to track the execution status of the standardized intervention strategy in the system and to verify the intervention results.

[0016] Fourthly, this application provides an electronic device, including a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; Memory, used to store computer programs; The processor, when executing a program stored in memory, implements the vehicle testing method as described in any of the first aspects.

[0017] Fifthly, this application provides a computer-readable storage medium storing a program for a vehicle testing method, wherein when the program for the vehicle testing method is executed by a processor, it implements the steps of the vehicle testing method as described in any of the first aspects.

[0018] The beneficial effects of this invention are: This application provides a vehicle testing and management method. After the cloud triggers the system to complete initialization, it sends a test task to the edge terminal, which then distributes it to the terminal. Upon receiving the test task, the terminal invokes a physical test cluster and a virtual multi-agent cluster to collaboratively execute the test task. During test execution, the entire test chain data is uploaded to the cloud in real time via the edge terminal. The edge terminal preprocesses and backs up the data locally before uploading it to the cloud in real time. The cloud receives the entire chain data and performs anomaly identification and risk warning level assessment based on it. If the risk warning level reaches the level required to trigger an intervention action, a corresponding standardized intervention strategy is matched according to the risk warning level, and the execution instruction of the standardized intervention strategy is sent to the terminal. Upon receiving the instruction, the terminal completes the corresponding handling action and triggers the entire chain fault recording. The cloud also tracks the system execution status of the standardized intervention strategy and verifies the intervention results. This application's solution is based on an integrated edge-cloud architecture, enabling control over the entire testing process of autonomous vehicles. It achieves intelligent control over the entire process, integrating real-time anomaly identification across all dimensions, virtual-real fusion scenario testing, cloud-based hierarchical real-time early warning, and automated closed-loop intervention. While meeting the requirements of testing efficiency and long cycle time, it also ensures the comprehensiveness and reliability of the testing. Attached Figure Description

[0019] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0020] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 A scenario architecture diagram of a vehicle testing control method as described in the application embodiment; Figure 2 This application provides a flowchart of a vehicle testing and control method; Figure 3 This application provides a schematic diagram of the process for cloud-based verification of intervention results in a vehicle testing and management method. Figure 4 This application provides a schematic diagram of the process for cloud-based verification of intervention results in another vehicle testing and control method. Figure 5 A flowchart of another vehicle testing and control method provided in this application embodiment; Figure 6 A schematic diagram of a vehicle testing and control device; Figure 7 This is a structural diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0023] In the field of intelligent connected vehicle autonomous driving technology, the testing requirements for intelligent vehicles are becoming increasingly demanding. As existing technologies still lack a complete technical solution that supports full-link system verification, covers the entire autonomous driving testing process, and integrates automated closed-loop intervention, this application provides a vehicle testing management method, system, device, electronic device, and storage medium.

[0024] The present application will now be described in detail through specific embodiments.

[0025] Figure 1 The following is a scenario architecture diagram of a vehicle testing control method as described in the application embodiment; firstly, referring to... Figure 1 As shown, the vehicle testing and management method provided in this application embodiment is implemented collaboratively by a three-layer architecture of cloud 100, edge 200 and terminal 300.

[0026] In this system architecture, Cloud100 serves as the core management layer, primarily responsible for the establishment, initiation, and configuration of test tasks (specifically, parsing test requirements, completing parameterized complex dynamic scenario modeling, configuring anomaly judgment criteria and graded early warning and intervention strategies), final analysis of test data (specifically, anomaly identification and risk level determination), triggering of graded early warnings and generation of intervention instructions, verification of test results, and system iteration (specifically, generating test reports, iterating virtual scenario libraries and AI anomaly identification models), while also undertaking the encrypted storage of data throughout the entire process.

[0027] The aforementioned edge terminal 200, as a collaborative adaptation layer, can be deployed at test intersections in actual implementation scenarios in the form of edge computing servers or edge computing gateways. It serves as a collaborative relay layer between the cloud and the terminal, and its function is to solve problems such as communication delay and data timing mismatch between the cloud and the terminal.

[0028] The edge terminal 300 receives configuration information and instructions from the cloud and forwards them to the terminal. It preprocesses and caches the raw test data uploaded by the terminal locally, verifies the execution status reported by the terminal and sends it back to the cloud. The edge terminal is also used to solve problems such as cloud-edge communication latency, virtual and real data timing mismatch, and network anomaly risks.

[0029] The aforementioned terminal 300, serving as the virtual-real test execution layer, includes physical vehicle testing equipment (such as physical vehicle testing units and roadside testing units) and virtual simulation multi-agent terminals in actual implementation scenarios. It enables bidirectional linkage between physical testing scenarios and virtual simulation scenarios, receives and executes start commands and intervention commands forwarded by the edge terminal (such as starting the test and completing intervention actions), collects full-link test data and packages and uploads it to the edge terminal, thereby achieving spatiotemporal alignment between the physical testing environment and the virtual simulation scenario.

[0030] In actual testing scenarios, the aforementioned terminal 300 can be the connected vehicle under test or a roadside sensing device, etc.

[0031] In this embodiment, a three-layer architecture is used to collaboratively manage the testing of the vehicle under test, which effectively solves the technical problems in the prior art, such as incomplete vehicle testing coverage, insufficient scenario adaptability, and lack of abnormal control and automated closed-loop handling in the testing process.

[0032] Figure 2 This application provides a flowchart of a vehicle testing and control method; see also... Figure 2 As shown, based on Figure 1 The architecture scenario shown in this application provides a vehicle testing method, including the following steps S10-S40: S10. After the triggering system completes initialization, the cloud sends the test task to the edge terminal, and then the edge terminal sends it to the terminal. After receiving the test task, the terminal calls the physical test cluster and the virtual multi-agent cluster to jointly execute the test task, and uploads the test full-link data to the cloud in real time through the edge terminal during the test execution.

[0033] In this embodiment, after the test is started, the physical test cluster of the terminal and the virtual multi-agent cluster perform collaborative testing according to the preset test cases, and synchronously collect the full-link node data of the intelligent driving system, the multi-agent interaction data, and the original data of vehicle-road collaboration status; wherein, the full-link node data includes the data generated by the entire link layer of the perception layer, decision layer, planning layer, control layer, multi-agent collaboration layer, and test process layer.

[0034] After the raw data undergoes preprocessing operations such as data cleaning, time alignment, and feature extraction at the edge, it is uploaded to the cloud in real time. At the same time, the edge performs a full local backup of the data to ensure data integrity.

[0035] S20. Receive the full-link data and perform anomaly identification and risk warning level judgment based on the full-link data.

[0036] S30. If the risk warning level reaches the level that triggers an intervention action, then a standardized intervention strategy corresponding to the risk warning level is matched, and the execution instruction of the standardized intervention strategy is sent to the terminal; so that the terminal receives the instruction, completes the corresponding handling action, and triggers full-link fault recording.

[0037] S40. The cloud tracking system tracks the execution status of the standardized intervention strategy and verifies the intervention results.

[0038] Figure 3 This is a schematic diagram illustrating the cloud-based verification of intervention results in a vehicle testing and control method provided in this application embodiment; see reference. Figure 3 As shown, the execution status of the standardized intervention strategy described in step S40 of the cloud tracking system is further verified to ensure the effectiveness of the intervention, including the following steps S401-403: S401. The cloud acquires the status data of the terminal during the execution of the intervention strategy in real time, and performs result verification based on the status data to determine whether the abnormality has been eliminated; S402. If so, resume the test process; S403. If not, continue to perform end-to-end anomaly identification and risk level assessment until the anomaly is eliminated and the testing process resumes.

[0039] Furthermore, in this embodiment of the application, anomaly identification and automatic intervention in abnormal situations are performed during the testing process of the vehicle under test, thereby improving the safety of the test.

[0040] Figure 5 A flowchart of another vehicle testing and control method provided in this application embodiment; see reference Figure 5 As shown in a specific embodiment of this application, the above method further includes the following step S50: Step S50: If the risk warning level does not reach the level to trigger an intervention action, then perform information recording, or push alarm information to the terminal and edge terminal, and save the scene recording when the anomaly occurs.

[0041] In one specific embodiment of this application, risk warning levels are pre-configured on the cloud side, including, for example, Level 1 warning, Level 2 warning, Level 3 warning and Level 4 warning; wherein, Level 1 warning corresponds to minor anomalies. In this state, measures for recording and logging are set up, without any additional active intervention actions. The abnormal information is only synchronously written to the distributed storage module for logging, which does not affect the normal execution of the test and does not require the terminal to perform any intervention operations.

[0042] Level 2 warning corresponds to a general abnormal state. In this state, the system is set to push alarm information to multiple terminals in real time and lock the scene recording when the abnormality occurs to retain the abnormal scene data, thereby ensuring that relevant test personnel pay attention and facilitating subsequent traceability and analysis. In this state, the test is not suspended and no additional actions are triggered on the terminal.

[0043] Level 3 warning corresponds to a more serious abnormal state. The corresponding intervention strategies under this state include: pushing emergency alarm information to multiple terminals in real time; suspending the current test process; synchronously writing abnormal information to the cloud distributed storage module for record keeping; generating standardized intervention instructions in the cloud and sending them down along the cloud-edge-terminal link; and executing corresponding actions after the terminal suspends the test operation.

[0044] Level 4 warning corresponds to a high-risk abnormal state. The intervention strategy under this state is as follows: push the highest level alarm information to multiple terminals in real time; write the abnormal information to the cloud distributed storage module for record keeping; trigger emergency safety intervention, generate standardized intervention instructions in the cloud, and send them down along the cloud-edge-terminal link. After the terminal suspends the test operation, it executes the corresponding disposal action (such as safely parking the vehicle) and triggers full-link fault recording.

[0045] Therefore, in this embodiment, during the intervention process of all levels of early warning, the early warning information and related data will be synchronously written into the distributed storage module for encrypted traces, providing a basis for subsequent anomaly analysis and system iteration optimization.

[0046] Figure 4 This is a schematic diagram illustrating the cloud-based verification of intervention results in another vehicle testing control method provided in this application embodiment; see reference. Figure 4 As shown, the method further includes: S404. If the anomaly is not eliminated, further determine whether the current risk level is the highest warning level; S405. If not, then upgrade to the highest risk warning level; S406. If so, then trigger the execution of the highest level security policy task.

[0047] Furthermore, in this embodiment, the cloud matches the standardized intervention strategy corresponding to the warning level and generates a disposal instruction that is sent down along the cloud-edge-terminal link; the terminal executes the corresponding intervention action and simultaneously triggers the full-link fault recording; the cloud verifies the system status after intervention in real time, and if the anomaly is eliminated, a test recovery instruction is issued; if the anomaly is not eliminated, the warning and intervention levels are automatically upgraded until the highest level of safety fallback action is triggered.

[0048] Furthermore, in the event of a network outage, the edge device automatically switches to the emergency mode, uses the local anomaly detection unit to detect high-risk anomalies, triggers local emergency intervention actions, and caches all data locally. Once the network is restored, the data is automatically re-uploaded to the cloud, ensuring test security and data integrity, and providing a safety net.

[0049] Referring to the specific scenarios of the risk level classification mentioned above, for example, if the anomaly is not eliminated, it is further determined whether the current risk level is a level four warning. If not, it is upgraded to a level four warning; if so, the highest level security policy task is triggered.

[0050] In this embodiment, automatic anomaly identification and hierarchical early warning are used to achieve anomaly gradient control, adapt to different risk scenarios, and provide corresponding differentiated intervention strategies to ensure test continuity and security. It is deeply adapted to the collaborative logic of the three-layer architecture to ensure that early warning and intervention are not interrupted, to ensure the closed loop of the entire test process, and to meet the needs of unattended testing.

[0051] In one specific embodiment of this application, before the triggering system completes the test environment initialization, the following steps A10-A20 are further included: Step A10: In response to the tester's test task creation operation, parse the test requirements and obtain the test case set; Step A20: Based on the test case set, perform the binding work between the terminal and the corresponding edge terminal, complete the parameterized modeling of the virtual scene and the configuration of test rules, and send the configuration information to the terminal and the edge terminal after compliance verification; This enables the terminal to achieve spatiotemporal alignment between the physical testing environment and the virtual simulation scene, and the edge terminal to achieve temporal synchronization and adaptation between the virtual and real scenes.

[0052] In this embodiment, during the test initialization phase, based on the tester's test task creation operation, the cloud parses the test requirements, obtains the test case set, generates test tasks, completes parameterized complex dynamic scenario modeling, and configures bidirectional mapping rules between the physical test cluster and the virtual multi-agent cluster, intelligent driving full-link anomaly judgment criteria, and hierarchical early warning and intervention strategy rules. After compliance verification, this configuration information is sent to the edge and terminal. At the edge, the timing synchronization and adaptation of the virtual and real scenarios are completed, and at the terminal, a one-to-one spatiotemporal alignment between the physical test cluster and the virtual multi-agent cluster scenario is achieved, completing the test environment initialization.

[0053] In a specific embodiment of this application, step S20, receiving the end-to-end data and performing anomaly identification and risk warning level judgment based on the end-to-end data, includes the following step C10: Step C10: Receive the full-link data uploaded by the terminal, covering the perception layer, decision-making layer, planning layer, control layer, multi-agent collaboration layer, and testing process layer, and complete the anomaly identification and risk level determination through a composite mechanism of threshold judgment, trend prediction, and AI anomaly identification.

[0054] For example, the cloud performs multi-dimensional parallel analysis on the uploaded test data, covering all types of anomalies in the perception layer, decision-making layer, planning layer, control layer, multi-agent collaboration layer, and test process layer. It completes accurate anomaly identification and risk level determination through a three-layer model of threshold judgment + trend prediction + AI anomaly recognition, triggering corresponding four-level warnings. Warning information is pushed to multiple terminals in real time, and the entire process data is encrypted, stored, and traced.

[0055] In one specific embodiment of this application, the cloud performs anomaly assessment by combining the YOLO series models based on deep learning with multi-target tracking technology and risk assessment indicators such as TTC (Time-to-Collision).

[0056] In a specific embodiment of this application, the above method further includes the following step D10: Step D10: After the testing is completed, the cloud platform generates a standardized test report and anomaly analysis report based on the full-process test data and pushes them to the cloud management backend.

[0057] Furthermore, in this embodiment, after the testing is completed, the cloud platform iteratively optimizes the virtual scene library based on the full-process test data, supplementing it with complex dynamic scene samples. Simultaneously, based on the newly added abnormal samples from this test, it incrementally trains and evaluates the accuracy of the AI ​​anomaly recognition model, and then releases the optimized model. This achieves self-optimization of system capabilities and completes the entire testing process loop.

[0058] The following is a detailed description of the specific implementation of the embodiments of this application through a concrete example: This embodiment uses the long-term reliability test of L4-level intelligent connected vehicles in a multi-vehicle cooperative traffic scenario at an urban intersection as the implementation scenario. The entire process relies on a three-layer architecture of cloud, edge, and terminal. The core objective is to verify the functional safety and long-term operational reliability of the tested L4-level intelligent driving domain control in vehicle-to-infrastructure (V2I) and multi-vehicle cooperative scenarios. The test cycle is 24 / 7 unattended automated testing. The test environment configuration includes: one standard urban intersection test vehicle, four tested connected vehicles equipped with L4-level intelligent driving domain control, eight sets of roadside perception and RSU (Roadside Unit) devices, and one virtual simulation platform. Specific implementation steps are illustrated below: Step 1: System deployment and communication link establishment; Terminal deployment (including physical terminals and virtual terminals): The physical terminal involved installing and debugging four connected vehicles and eight roadside sensing and RSU devices at the test intersection, deploying data acquisition and command execution nodes to ensure data acquisition without loss and command execution smoothly, and enabling the roadside devices to collect environmental and traffic-related data normally. The virtual terminal involved deploying a virtual simulation terminal to build a simulation environment with virtual background vehicles and virtual roadside and traffic signal facilities, completing the communication configuration between the physical and virtual terminals, and realizing two-way linkage between the virtual and real scenarios.

[0059] Edge deployment: Deploy an edge gateway near the test intersection, complete the gigabit Ethernet communication configuration with the terminal layer, deploy components such as data preprocessing, command forwarding, local caching, and network outage emergency response, configure basic operating rules, and ensure efficient data transmission and command forwarding.

[0060] Cloud deployment: Build a visual management and control backend, a distributed storage cluster and a real-time computing engine, complete the construction of encrypted communication links with the edge, configure user permissions and scheduling rules, and ensure normal communication of the three-layer architecture.

[0061] Step 2: Test task configuration and virtual-real scene mapping; Test task creation: Test engineers complete identity verification through the cloud, create 24 / 7 long-cycle test tasks, bind test equipment, upload test case sets for multi-vehicle collaboration at city intersections, configure the loop duration for single scenarios and full test cases, and set four-level warning suspension rules for testing.

[0062] Scene modeling and virtual-physical mapping: Parametric test scene modeling is completed through the cloud, and bidirectional mapping rules between virtual and physical scenes are configured to ensure 1:1 spatiotemporal alignment between virtual and physical scenes and real-time data synchronization, thus guaranteeing scene fidelity and data authenticity.

[0063] Rule configuration: Configure the intelligent driving full-link anomaly judgment rules, four-level graded early warning strategy and automated intervention rules. After all configurations are verified for compliance, they are saved and distributed to the edge terminal and the terminal.

[0064] Step 3: Initialize the test environment and transmit status data. Edge initialization: Receive configuration information from the cloud, complete the synchronization and adaptation of virtual and real scenarios, and send initialization instructions to the terminal.

[0065] Terminal initialization: The physical and virtual terminals are scheduled to complete the initialization, including the startup of the real vehicle and roadside equipment, intelligent driving domain control self-test, virtual scene loading, and data acquisition unit startup.

[0066] Status feedback and confirmation: The terminal summarizes the ready status, verifies it at the edge, and then sends it back to the cloud. After the test engineer confirms that there are no abnormalities, the test start command is issued.

[0067] Step 4: Test execution and end-to-end data flow; Test execution: The test task start command is sent to the terminal via the cloud and edge. The terminal starts the automated test sequence and executes the test in a loop according to the test cases. The real vehicle and the virtual vehicle form a multi-agent collaborative scenario. The intelligent driving domain control executes algorithms based on various data to form a vehicle-road-cloud test closed loop.

[0068] Terminal data collection and uploading: The terminal collects intelligent driving data in real time, packages and summarizes it, and then uploads it to the edge terminal.

[0069] Edge data processing and uploading: Preprocessing of data uploaded from terminals, such as cleaning and time-series alignment, is performed before uploading to the cloud in real time, while local caching and backup are also performed to ensure data integrity.

[0070] Step 5: Anomaly identification, tiered early warning, and automated closed-loop intervention; Anomaly identification and classification: Real-time cloud-based analysis of test data, using a three-layer model to complete full-dimensional anomaly identification and risk level determination from level one to level four.

[0071] Tiered warning triggering: Level 1 warning only records and leaves a trace; Level 2 warning pushes an alarm and locks the scene for recording; Level 3 warning pushes an emergency alarm and suspends testing; Level 4 warning pushes the highest alarm and triggers emergency security intervention; all warning information is pushed to multiple devices and encrypted and left a trace.

[0072] Automated closed-loop intervention: The cloud generates intervention instructions that match the warning level, forwards them to the terminal for execution via the edge, the terminal provides feedback on the execution status, and the cloud verifies whether the anomaly has been eliminated. If it has not been eliminated, the warning is upgraded until the highest level of intervention is triggered.

[0073] Emergency response to network outage: When the network is interrupted, the edge device switches to emergency mode, monitors high-risk anomalies and triggers local intervention, caches data, and re-uploads it to the cloud after the network is restored.

[0074] Step 6: Final testing and system capability iteration; Test completion: After the test task is completed, the terminal stops the test and summarizes the status. After verification by the edge terminal, the data is sent back to the cloud. The edge terminal then synchronizes the cached data to the cloud.

[0075] Report generation: The cloud retrieves data from the entire process and automatically generates standardized test reports and anomaly analysis reports, which are then pushed to relevant terminals for viewing, downloading, and archiving.

[0076] System iteration: Optimize the virtual scene library based on test data, incrementally train the AI ​​anomaly recognition model and evaluate its accuracy, and synchronize the optimized model to each layer to provide support for subsequent testing.

[0077] Furthermore, the entire system achieves intelligent control over the entire process of autonomous vehicle testing, integrating real-time anomaly identification across all dimensions, virtual-real fusion scenario testing, cloud-based hierarchical real-time early warning, and automated closed-loop intervention.

[0078] One specific embodiment of this application provides a vehicle testing and management system, which is based on the above-mentioned three-layer architecture, including: cloud, edge and terminal.

[0079] The aforementioned cloud is used to send test tasks to the edge terminal after the trigger system completes initialization, and then distribute the tasks to the terminal via the edge terminal. The aforementioned terminal is used to, upon receiving the test task, invoke the physical test cluster and the virtual multi-agent cluster to collaboratively execute the test work, and send the test end-to-end data to the edge terminal in real time during the test execution process; After preprocessing and backing up the data locally, the aforementioned edge devices upload the data to the cloud in real time. The aforementioned cloud platform is also used to receive the end-to-end data and to perform anomaly identification and risk warning level judgment based on the end-to-end data; If the risk warning level reaches the level that triggers an intervention action, a corresponding standardized intervention strategy is matched according to the risk warning level, and the execution instruction of the standardized intervention strategy is sent to the terminal; so that the terminal receives the instruction, completes the corresponding handling action, and triggers full-link fault recording. The system tracks the execution status of the standardized intervention strategy and verifies the intervention results.

[0080] The specific workflow of the vehicle testing and control system provided in this application embodiment is described in the above-described embodiment of the vehicle control method; it will not be repeated here.

[0081] The vehicle test management system provided in this application solves the problems of incomplete test coverage, weak scenario adaptability, and inability to monitor and intervene in real time in the cloud by constructing a test anomaly real-time early warning and intervention technical solution based on an end-edge-cloud integrated architecture, covering the entire intelligent driving link, integrating virtual and real scene mapping, and supporting multi-agent collaboration. It achieves the effect of automatic anomaly control and automated closed loop in the entire collaborative test process.

[0082] Figure 6 This is a schematic diagram of the structure of a vehicle testing and control device provided in an embodiment of this application; see reference. Figure 6 As shown, the vehicle testing apparatus includes: The test task sending module 601 is used to send test tasks from the cloud to the edge terminal after the trigger system completes initialization, and then distribute the test tasks to the terminal via the edge terminal. This allows the terminal to receive the test tasks, invoke the physical test cluster and the virtual multi-agent cluster to collaboratively execute the test tasks, and upload the test data of the entire test link to the cloud in real time via the edge terminal during the test execution process. The edge terminal performs preprocessing and local backup of the data before uploading it to the cloud in real time. Anomaly identification module 602 is used to receive the end-to-end data and perform anomaly identification and risk warning level judgment based on the end-to-end data; The intervention strategy matching module 603 is used to match a standardized intervention strategy according to the risk warning level if the risk warning level reaches the level that triggers an intervention action, and send the execution instruction of the standardized intervention strategy to the terminal; so that the terminal can complete the corresponding handling action after receiving the instruction and trigger the full-link fault recording. The verification module 604 is used to track the execution status of the standardized intervention strategy of the system and verify the intervention results.

[0083] In a specific embodiment of this application, the verification module 604 is specifically used to include: The cloud acquires execution status data in real time during the execution of the intervention strategy by the terminal, and performs result verification based on the status data to determine whether the anomaly has been eliminated; If so, resume the testing process; If not, continue with end-to-end anomaly identification and risk level assessment until the anomaly is eliminated and the testing process resumes.

[0084] In one specific embodiment of this application, the above-mentioned device further includes: The test task creation module (not shown in the figure) is used to respond to the tester's test task creation operation, parse the test requirements, and obtain the test case set; The configuration module is used to perform the binding work between the terminal and the corresponding edge terminal according to the test case set, complete the parameterized modeling of the virtual scene and the configuration of test rules, and send the configuration information to the terminal and the edge terminal after compliance verification. This enables the terminal to achieve spatiotemporal alignment between the physical testing environment and the virtual simulation scene, and the edge device to achieve temporal synchronization and adaptation between the virtual and real scenes.

[0085] In one specific embodiment of this application, the verification module 604 is further configured to: If the risk warning level does not reach the level that triggers intervention, then information recording is performed, or alarm information is pushed to the terminal and edge terminal, and the scene recording when the anomaly occurs is saved.

[0086] In a specific embodiment of this application, the anomaly identification module 602 is specifically used for: The system receives end-to-end data from the receiving terminal, covering the perception layer, decision-making layer, planning layer, control layer, multi-agent collaboration layer, and testing process layer. It then uses a composite mechanism of threshold judgment, trend prediction, and AI anomaly recognition to identify anomalies and determine risk levels.

[0087] In one specific embodiment of this application, the above-mentioned device further includes: The report generation module is used to generate standardized test reports and anomaly analysis reports based on the full-process test data after the testing work is completed in the cloud, and push them to the cloud management backend.

[0088] In one specific embodiment of this application, the above-mentioned device further includes: The optimization module (not shown in the figure) is used to perform iterative optimization of the virtual scene library based on the full-process test data after the testing work is completed, and to supplement complex dynamic scene samples. Based on the newly added abnormal samples in this test, the AI ​​anomaly recognition model will be incrementally trained and its accuracy evaluated, and the optimized model will be released.

[0089] Figure 7 A structural diagram of an electronic device provided in an embodiment of this application; see reference Figure 7 As shown in the embodiments of this application, an electronic device is provided, including a processor 1110, a communication interface 1120, a memory 1130 and a communication bus 1140, wherein the processor 1110, the communication interface 1120 and the memory 1130 communicate with each other through the communication bus. Memory 1130 is used to store computer programs; The processor 1110, when executing a program stored in the memory, implements the vehicle test monitoring method described in any of the foregoing method embodiments, including: after the trigger system completes initialization, the cloud sends a test task to the edge terminal, and the edge terminal distributes it to the terminal, so that after receiving the test task, the terminal calls the physical test cluster and the virtual multi-agent cluster to collaboratively execute the test task, and uploads the test full-link data to the cloud in real time via the edge terminal during the test execution process; wherein, the edge terminal preprocesses the data and uploads it to the cloud in real time, while simultaneously performing local backup; Receive the end-to-end data and perform anomaly identification and risk warning level determination based on the end-to-end data; If the risk warning level reaches the level that triggers an intervention action, a corresponding standardized intervention strategy is matched according to the risk warning level, and the execution instruction of the standardized intervention strategy is sent to the terminal; so that the terminal receives the instruction, completes the corresponding handling action, and triggers full-link fault recording. The system tracks the execution status of the standardized intervention strategy and verifies the intervention results.

[0090] In one possible implementation, the tracking system monitors the execution status of the standardized intervention strategy and verifies the intervention results, including: The cloud acquires execution status data in real time during the execution of the intervention strategy by the terminal, and performs result verification based on the status data to determine whether the anomaly has been eliminated; If so, resume the testing process; If not, continue with end-to-end anomaly identification and risk level assessment until the anomaly is eliminated and the testing process resumes.

[0091] In one possible implementation, before the triggering system completes the test environment initialization, the following is included: In response to the tester's test task creation operation, the test requirements are parsed and the test case set is obtained; Based on the test case set, the binding work between the terminal and the corresponding edge terminal is performed, the parameterized modeling of the virtual scene and the configuration of test rules are completed, and the configuration information is sent to the terminal and the edge terminal after compliance verification. This enables the terminal to achieve spatiotemporal alignment between the physical testing environment and the virtual simulation scene, and the edge device to achieve temporal synchronization and adaptation between the virtual and real scenes.

[0092] In one possible implementation, the method further includes: If the risk warning level does not reach the level that triggers intervention, then information recording is performed, or alarm information is pushed to the terminal and edge terminal, and the scene recording when the anomaly occurs is saved.

[0093] In one possible implementation, receiving the end-to-end data and performing anomaly identification and risk warning level determination based on the end-to-end data includes: The system receives end-to-end data from the receiving terminal, covering the perception layer, decision-making layer, planning layer, control layer, multi-agent collaboration layer, and testing process layer. It then uses a composite mechanism of threshold judgment, trend prediction, and AI anomaly recognition to identify anomalies and determine risk levels.

[0094] In one possible implementation, the method further includes: After the testing is completed, the cloud platform generates standardized test reports and anomaly analysis reports based on the full-process test data and pushes them to the cloud management backend.

[0095] In one possible implementation, the method further includes: After the testing is completed, the cloud platform uses the full-process test data to iteratively optimize the virtual scene library and supplement it with complex dynamic scene samples. Based on the newly added abnormal samples in this test, the AI ​​anomaly recognition model will be incrementally trained and its accuracy evaluated, and the optimized model will be released.

[0096] In one possible implementation, the method further includes: If the anomaly is not eliminated, further determine whether the current risk level is the highest warning level; If not, then upgrade to the highest risk warning level; If so, the task of executing the highest level of security policy will be triggered.

[0097] The electronic device provided in this embodiment of the invention includes a processor 1110 that executes a program stored in its memory. After the cloud triggers system initialization, it sends a test task to the edge terminal, which then distributes it to the terminal. Upon receiving the test task, the terminal invokes a physical test cluster and a virtual multi-agent cluster to collaboratively execute the test task. During test execution, the entire test chain data is uploaded to the cloud in real time via the edge terminal. The edge terminal preprocesses and backs up the data locally before uploading it to the cloud. The cloud receives the entire chain data and performs anomaly identification and risk warning level assessment based on it. If the risk warning level reaches the level required to trigger an intervention action, a corresponding standardized intervention strategy is matched according to the risk warning level, and the execution instruction of the standardized intervention strategy is sent to the terminal. Upon receiving the instruction, the terminal completes the corresponding handling action and triggers the entire chain fault recording. Furthermore, the cloud tracks the system execution status of the standardized intervention strategy and verifies the intervention results.

[0098] This application's solution is based on an integrated edge-cloud architecture, enabling full-process control of autonomous vehicle testing. It achieves intelligent full-process control that integrates real-time identification of anomalies across all dimensions, testing in virtual and real scenarios, real-time cloud-based hierarchical early warning, and automated closed-loop intervention. While meeting the requirements of testing efficiency and long cycle time, it also ensures the comprehensiveness and reliability of the testing.

[0099] The communication bus 1140 mentioned in the above-mentioned electronic device can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This communication bus 1140 can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 7 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0100] The communication interface 1120 is used for communication between the above-mentioned electronic device and other devices.

[0101] The memory 1130 may include random access memory (RAM) or non-volatile memory, such as at least one disk storage device. Optionally, the memory may also be at least one storage device located remotely from the aforementioned processor.

[0102] The processor 1110 mentioned above can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0103] In another embodiment of this application, a computer-readable storage medium is provided, on which a program for a vehicle testing and control method is stored. When executed by a processor, the program for the vehicle testing and control method implements the steps of the vehicle testing and control method described in any of the foregoing method embodiments, including: After the trigger system completes initialization, the cloud sends test tasks to the edge terminal, which then distributes them to the terminal. Upon receiving the test task, the terminal invokes the physical test cluster and the virtual multi-agent cluster to collaboratively execute the test task. During the test execution, the entire test chain data is uploaded to the cloud in real time via the edge terminal. The edge terminal preprocesses the data before uploading it to the cloud in real time, while simultaneously performing local backups. Receive the end-to-end data and perform anomaly identification and risk warning level determination based on the end-to-end data; If the risk warning level reaches the level that triggers an intervention action, a corresponding standardized intervention strategy is matched according to the risk warning level, and the execution instruction of the standardized intervention strategy is sent to the terminal; so that the terminal receives the instruction, completes the corresponding handling action, and triggers full-link fault recording. The system tracks the execution status of the standardized intervention strategy and verifies the intervention results.

[0104] In one possible implementation, the tracking system monitors the execution status of the standardized intervention strategy and verifies the intervention results, including: The cloud acquires execution status data in real time during the execution of the intervention strategy by the terminal, and performs result verification based on the status data to determine whether the anomaly has been eliminated; If so, resume the testing process; If not, continue with end-to-end anomaly identification and risk level assessment until the anomaly is eliminated and the testing process resumes.

[0105] In one possible implementation, before the triggering system completes the test environment initialization, the following is included: In response to the tester's test task creation operation, the test requirements are parsed and the test case set is obtained; Based on the test case set, the binding work between the terminal and the corresponding edge terminal is performed, the parameterized modeling of the virtual scene and the configuration of test rules are completed, and the configuration information is sent to the terminal and the edge terminal after compliance verification. This enables the terminal to achieve spatiotemporal alignment between the physical testing environment and the virtual simulation scene, and the edge device to achieve temporal synchronization and adaptation between the virtual and real scenes.

[0106] In one possible implementation, the method further includes: If the risk warning level does not reach the level that triggers intervention, then information recording is performed, or alarm information is pushed to the terminal and edge terminal, and the scene recording when the anomaly occurs is saved.

[0107] In one possible implementation, receiving the end-to-end data and performing anomaly identification and risk warning level determination based on the end-to-end data includes: The system receives end-to-end data from the receiving terminal, covering the perception layer, decision-making layer, planning layer, control layer, multi-agent collaboration layer, and testing process layer. It then uses a composite mechanism of threshold judgment, trend prediction, and AI anomaly recognition to identify anomalies and determine risk levels.

[0108] In one possible implementation, the method further includes: After the testing is completed, the cloud platform generates standardized test reports and anomaly analysis reports based on the full-process test data and pushes them to the cloud management backend.

[0109] In one possible implementation, the method further includes: After the testing is completed, the cloud platform uses the full-process test data to iteratively optimize the virtual scene library and supplement it with complex dynamic scene samples. Based on the newly added abnormal samples in this test, the AI ​​anomaly recognition model will be incrementally trained and its accuracy evaluated, and the optimized model will be released.

[0110] In one possible implementation, the method further includes: If the anomaly is not eliminated, further determine whether the current risk level is the highest warning level; If not, then upgrade to the highest risk warning level; If so, the task of executing the highest level of security policy will be triggered.

[0111] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0112] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features claimed herein.

Claims

1. A vehicle testing control method, characterized in that, The method is applied in the cloud and includes: After the trigger system completes initialization, the cloud sends test tasks to the edge terminal, which then distributes them to the terminal. Upon receiving the test task, the terminal invokes the physical test cluster and the virtual multi-agent cluster to collaboratively execute the test task. During the test execution, the entire test chain data is uploaded to the cloud in real time via the edge terminal. The edge terminal preprocesses the data before uploading it to the cloud in real time, while simultaneously performing local backups. Receive the end-to-end data and perform anomaly identification and risk warning level determination based on the end-to-end data; If the risk warning level reaches the level that triggers an intervention action, a corresponding standardized intervention strategy is matched according to the risk warning level, and the execution instruction of the standardized intervention strategy is sent to the terminal; so that the terminal receives the instruction, completes the corresponding handling action, and triggers full-link fault recording. The system tracks the execution status of the standardized intervention strategy and verifies the intervention results.

2. The method according to claim 1, characterized in that, The tracking system monitors the execution status of the standardized intervention strategy and verifies the intervention results, including: The cloud acquires execution status data in real time during the execution of the intervention strategy by the terminal, and performs result verification based on the status data to determine whether the anomaly has been eliminated; If so, resume the testing process; If not, continue with end-to-end anomaly identification and risk level assessment until the anomaly is eliminated and the testing process resumes.

3. The method according to claim 1, characterized in that, Before the triggering system completes the test environment initialization, the following is included: In response to the tester's test task creation operation, the test requirements are parsed and the test case set is obtained; Based on the test case set, the binding work between the terminal and the corresponding edge terminal is performed, the parameterized modeling of the virtual scene and the configuration of test rules are completed, and the configuration information is sent to the terminal and the edge terminal after compliance verification. This enables the terminal to achieve spatiotemporal alignment between the physical testing environment and the virtual simulation scene, and the edge device to achieve temporal synchronization and adaptation between the virtual and real scenes.

4. The method according to claim 1, characterized in that, The method further includes: If the risk warning level does not reach the level that triggers intervention, then information recording is performed, or alarm information is pushed to the terminal and edge terminal, and the scene recording when the anomaly occurs is saved.

5. The method according to claim 1, characterized in that, The process of receiving the end-to-end data and performing anomaly identification and risk warning level determination based on the end-to-end data includes: The system receives end-to-end data from the receiving terminal, covering the perception layer, decision-making layer, planning layer, control layer, multi-agent collaboration layer, and testing process layer. It then uses a composite mechanism of threshold judgment, trend prediction, and AI anomaly recognition to identify anomalies and determine risk levels.

6. The method according to claim 1, characterized in that, The method further includes: After the testing is completed, the cloud platform generates standardized test reports and anomaly analysis reports based on the full-process test data and pushes them to the cloud management backend.

7. The method according to claim 5, characterized in that, The method further includes: After the testing is completed, the cloud platform uses the full-process test data to iteratively optimize the virtual scene library and supplement it with complex dynamic scene samples. Based on the newly added abnormal samples in this test, the AI ​​anomaly recognition model will be incrementally trained and its accuracy evaluated, and the optimized model will be released.

8. The method according to claim 2, characterized in that, The method further includes: If the anomaly is not eliminated, further determine whether the current risk level is the highest warning level; If not, then upgrade to the highest risk warning level; If so, the task of executing the highest level of security policy will be triggered.

9. A vehicle testing and control system, characterized in that, The system includes: cloud, edge, and terminal; The cloud is used to send test tasks to the edge terminal after the trigger system completes initialization, and then distribute the tasks to the terminal via the edge terminal. The terminal is used to receive the test task, call the physical test cluster and the virtual multi-agent cluster to jointly execute the test work, and send the test full-link data to the edge terminal in real time during the test execution process; The edge device preprocesses and backs up the data locally, then uploads it to the cloud in real time. The cloud platform is also used to receive the end-to-end data and to perform anomaly identification and risk warning level judgment based on the end-to-end data. If the risk warning level reaches the level that triggers an intervention action, a corresponding standardized intervention strategy is matched according to the risk warning level, and the execution instruction of the standardized intervention strategy is sent to the terminal; so that the terminal receives the instruction, completes the corresponding handling action, and triggers full-link fault recording. The system tracks the execution status of the standardized intervention strategy and verifies the intervention results.

10. A vehicle testing and control device, characterized in that, include: The test task sending module is used to send test tasks from the cloud to the edge terminal after the trigger system completes initialization, and then distribute the test tasks to the terminal via the edge terminal. After receiving the test task, the terminal calls the physical test cluster and the virtual multi-agent cluster to collaboratively execute the test task, and uploads the test end-to-end data to the cloud in real time via the edge terminal during the test execution. The edge terminal preprocesses the data and uploads it to the cloud in real time, while also performing local backup. Anomaly detection module is used to receive the end-to-end data and perform anomaly detection and risk warning level judgment based on the end-to-end data; An intervention strategy matching module is used to match a standardized intervention strategy according to the risk warning level if the risk warning level reaches the level that triggers an intervention action, and send the execution instruction of the standardized intervention strategy to the terminal; so that the terminal can complete the corresponding handling action after receiving the instruction and trigger the end-to-end fault recording. The verification module is used to track the execution status of the standardized intervention strategy in the system and to verify the intervention results.

11. An electronic device, characterized in that, It includes a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; Memory, used to store computer programs; The processor, when executing a program stored in memory, implements the vehicle testing and control method according to any one of claims 1 to 9.

12. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a program for a vehicle testing and control method, which, when executed by a processor, implements the steps of the vehicle testing and control method according to any one of claims 1-9.

Images