Automatic driving test method and device, electronic equipment and storage medium

Abstract

This application provides an autonomous driving testing method, apparatus, electronic device, and storage medium, relating to the field of intelligent driving technology. The method includes: collecting real-vehicle test data through a front-end device, the real-vehicle test data including a set of parameter signals characterizing the autonomous driving conditions of the vehicle during real-vehicle testing; determining whether there are anomalies in the real-vehicle test data based on a preset detection strategy; when anomalies are found in the real-vehicle test data, identifying problem points in the domain control software based on the real-vehicle test data, and using these problem points as test results, wherein the problem points include the control modules corresponding to the parameter signals causing the anomalies. This can improve the problem of high test bench costs and cumbersome test environment setup and management caused by multiple real-vehicle tests due to the complexity of autonomous driving testing requirements.

Description

Technical Field

[0001] This invention relates to the field of intelligent driving technology, and more specifically, to an autonomous driving testing method, apparatus, electronic device, and storage medium. Background Technology

[0002] With the development of autonomous driving technology, vehicles with autonomous driving capabilities have become commonplace in households. Currently, vehicles with autonomous driving capabilities typically undergo three testing phases before commercialization: simulation testing, closed-track testing, and open-road testing. Among these, simulation testing is a crucial step in realizing the practical application of advanced autonomous driving. Autonomous driving simulation testing primarily uses mathematical modeling to digitally recreate autonomous driving application scenarios, establishing a system model that closely resembles the real world. This allows for the testing and verification of autonomous driving systems and algorithms directly through software simulation testing without the need for actual vehicles.

[0003] Existing autonomous driving testing methods often require the use of test benches, which are highly customized. If multiple tests are needed, multiple purchases of test benches are necessary. Therefore, existing autonomous driving testing methods suffer from high costs and cumbersome setup and management of multi-scenario testing environments. Summary of the Invention

[0004] In view of this, the purpose of this application is to provide an autonomous driving testing method, device, electronic device and storage medium, which can improve the problem of high test bench costs and complicated test environment construction and management caused by multiple real vehicle tests due to the complexity of autonomous driving testing requirements.

[0005] To achieve the above technical objectives, the technical solution adopted in this application is as follows:

[0006] In a first aspect, embodiments of this application provide an autonomous driving testing method, the method comprising:

[0007] Real-vehicle test data is collected through front-end devices. The real-vehicle test data includes a set of parameter signals used to characterize the autonomous driving conditions of the vehicle in the real-vehicle test.

[0008] Based on a preset detection strategy, determine whether there are any anomalies in the actual vehicle test data;

[0009] When there are abnormalities in the actual vehicle test data, the problem points in the domain control software are determined based on the actual vehicle test data, and the problem points are used as test results. The problem points include the control modules corresponding to the parameter signals that caused the abnormalities.

[0010] In conjunction with the first aspect, in some alternative implementations, the method further includes:

[0011] When the domain controller software is initially identified as having the aforementioned problem, a preset data backflow strategy is executed.

[0012] When the domain controller software generates incorrect control commands or fails to respond based on the data backfeed strategy, the problem point in the domain controller software is determined to be valid.

[0013] In conjunction with the first aspect, in some optional implementations, a preset data backfeeding strategy is executed, including:

[0014] A vehicle simulation test environment is generated based on the real vehicle test data. The vehicle simulation test environment is used to characterize the normal autonomous driving conditions of the vehicle in the simulation test.

[0015] The domain control software is run in the vehicle simulation test environment;

[0016] The domain control software executes a corresponding test script based on the problem point. The test script carries environmental parameters, which are used to characterize the road environment features when the problem point occurs.

[0017] In conjunction with the first aspect, in some optional implementations, before determining the problem points in the domain control software based on the actual vehicle test data, the method further includes:

[0018] The actual vehicle test data is backed up as a .dat file.

[0019] In conjunction with the first aspect, in some alternative implementations, the method further includes:

[0020] The domain controller software will be iteratively updated based on the identified problems.

[0021] Based on the updated domain controller software, a preset regression verification strategy is executed;

[0022] When the domain controller software generates the correct control command based on the regression verification strategy, the repair result of the problem point is determined to be effective.

[0023] In conjunction with the first aspect, in some optional implementations, a preset regression verification strategy is executed, including:

[0024] A vehicle simulation test environment is generated based on the actual vehicle test data.

[0025] The updated domain control software is run in the vehicle simulation test environment, wherein the updated domain control software is used to execute corresponding test scripts based on the problem points.

[0026] In conjunction with the first aspect, in some optional implementations, based on a preset detection strategy, determining whether the real vehicle test data is abnormal includes:

[0027] When the front-end device senses that the lane environment meets the preset vehicle lane-changing conditions, and the domain control software cannot control the vehicle to perform the corresponding lane-changing operation according to the vehicle lane-changing conditions, it is determined that the real vehicle test data is abnormal.

[0028] When the front-end device senses that the distance between the vehicle and the solid lines on both sides of the lane is not equal, and the domain control software cannot control the vehicle to return to the center position of the lane, it is determined that there is an anomaly in the real vehicle test data.

[0029] When the front-end device senses that the vehicle deviates from the preset planned route, and the domain control software is unable to replan the route between the current vehicle position and the destination, it determines that the real vehicle test data is abnormal.

[0030] Secondly, embodiments of this application also provide an autonomous driving testing device, the device comprising:

[0031] The acquisition unit is used to acquire real vehicle test data of the vehicle through front-end equipment. The real vehicle test data includes a set of parameter signals used to characterize the autonomous driving conditions of the vehicle in the real vehicle test.

[0032] The judgment unit is used to determine whether there are any abnormalities in the actual vehicle test data based on a preset detection strategy;

[0033] The determining unit is used to determine the problem points in the domain control software based on the actual vehicle test data when there are abnormalities in the actual vehicle test data, and to take the problem points as test results. The problem points include the control modules corresponding to the parameter signals that caused the abnormalities.

[0034] Thirdly, embodiments of this application also provide an electronic device, which includes a processor and a memory coupled to each other, wherein a computer program is stored in the memory, and when the computer program is executed by the processor, the electronic device performs the above-described method.

[0035] Fourthly, embodiments of this application also provide a computer-readable storage medium storing a computer program that, when run on a computer, causes the computer to perform the methods described above.

[0036] The invention employing the above technical solution has the following advantages:

[0037] In the technical solution provided in this application, real-vehicle test data is collected by a front-end device deployed on the vehicle. A detection strategy is then used to determine if any anomalies exist in the real-vehicle test data. When anomalies are found, the corresponding control module in the domain control software is identified as the problem point causing the anomaly, and this problem point is used as the test result. This facilitates the accurate location of problems in the domain control software and avoids multiple real-vehicle tests due to the complexity of autonomous driving testing requirements by using the domain control software as a repeatedly testable object. This reduces bench costs and simplifies testing procedures. Attached Figure Description

[0038] This application can be further illustrated by the non-limiting embodiments given in the accompanying drawings. It should be understood that the following drawings only illustrate some embodiments of this application and should not be considered as limiting the scope. For those skilled in the art, other related drawings can be obtained from these drawings without any inventive effort.

[0039] Figure 1 A block diagram of an electronic device provided in an embodiment of this application.

[0040] Figure 2 This is a flowchart illustrating the autonomous driving testing method provided in an embodiment of this application.

[0041] Figure 3 A block diagram of an autonomous driving test device provided in an embodiment of this application.

[0042] Icons: 100 - Electronic device; 101 - Processor; 102 - Memory; 200 - Autonomous driving test device; 210 - Data acquisition unit; 220 - Judgment unit; 230 - First determination unit. Detailed Implementation

[0043] The present application will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that similar or identical parts are referred to by the same reference numerals in the drawings or description. Implementations not shown or described in the drawings are forms known to those skilled in the art. In the description of this application, terms such as "first" and "second" are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0044] Please refer to Figure 1 This application provides an electronic device 100 that may include a processor 101 and a memory 102. The memory 102 stores a computer program, which, when executed by the processor 101, enables the electronic device 100 to perform the corresponding steps in the following autonomous driving test method.

[0045] Please refer to Figure 2This application provides an autonomous driving testing method. The autonomous driving testing method may include the following steps:

[0046] Step 110: Collect real-vehicle test data through front-end devices;

[0047] Step 120: Based on a preset detection strategy, determine whether there are any anomalies in the actual vehicle test data;

[0048] Step 130: When there are abnormalities in the actual vehicle test data, determine the problem points in the domain control software based on the actual vehicle test data, and use the problem points as the test results.

[0049] In the above implementation, real-vehicle test data is collected by front-end devices deployed on the vehicle. A detection strategy is then used to determine if any anomalies exist in the real-vehicle test data. When anomalies are found, the corresponding control module in the domain control software is identified as the problem point causing the anomaly, and this problem point is used as the test result. This facilitates the precise location of problems in the domain control software and avoids multiple real-vehicle tests due to the complexity of autonomous driving testing requirements by using the domain control software as a repeatedly testable object. This reduces bench costs and simplifies the testing process.

[0050] The steps of the autonomous driving testing method will be explained in detail below:

[0051] In step 110, the real vehicle test data may include a set of parameter signals used to characterize the autonomous driving conditions of the vehicle in the real vehicle test. The set of parameter signals may include multiple parameter signals used to characterize the autonomous driving conditions of the vehicle in the real vehicle test. The parameter signals may be positioning signals, radar signals, lane change signals, vehicle control signals, lateral and longitudinal status signals, trajectory planning signals, etc.

[0052] In this embodiment, the front-end device can be a camera, radar, GPS, etc. The front-end device can be deployed in a distributed manner on the vehicle, or it can be integrated as a standalone device and installed on the vehicle, and the front-end device is electrically connected to the vehicle.

[0053] In step 120, the preset detection strategy can be understood as a judgment condition. For example, during autonomous driving, a traffic sign indicating a right turn appears in the lane, at which point the vehicle needs to change lanes to enter the right turn lane. When the vehicle correctly changes lanes and turns right based on the right turn sign, it is determined that the real vehicle test data is normal; when the vehicle detects the right turn sign through the front-end device, but the domain control software in the vehicle controls the vehicle to change lanes to the left lane based on the right turn sign, or when the domain control software does not respond, it is determined that the real vehicle test data is abnormal.

[0054] In step 130, the problem point may include the control module corresponding to the parameter signal that caused the abnormal situation.

[0055] In this embodiment, the domain control software is the conventional control software for vehicles with autonomous driving capabilities, used to control the vehicle to perform corresponding driving operations based on the lane environment or preset programs.

[0056] In this embodiment, the domain control software includes one or more virtual control modules. Each control module corresponds to one or more front-end devices. The control module generates corresponding control commands based on the parameter signals sent to the domain control software by the front-end devices and controls the vehicle to perform corresponding driving operations. For example, the domain control software has a lane-changing control module. When the vehicle is in autonomous driving mode, and the radar speed detector deployed on the vehicle detects that the vehicle in front is moving slowly, and the camera deployed on the vehicle detects that the left lane is clear and empty, the lane-changing control module receives the parameter signals from the radar speed detector and the camera, and generates a control command for the vehicle to change lanes to the left, thereby controlling the vehicle to change lanes to the left to overtake.

[0057] In this embodiment, when an anomaly is determined in the real-vehicle test data, the control module corresponding to the anomaly is identified as the problem point. For example, the domain control software has a lane-changing control module. When the vehicle is traveling in the middle lane of a three-lane road, and the intersection ahead is a right-turn intersection according to the vehicle's preset trajectory, the camera deployed on the vehicle detects a right-turn traffic sign on the right lane and that the right lane is clear. The lane-changing control module receives the parameter signal from the camera but does not generate a control command for the vehicle to change lanes to the right; the vehicle remains in a straight-ahead state. That is, the vehicle exhibits an anomaly where it cannot automatically change lanes based on the road environment, and the lane-changing control module corresponding to this anomaly is the problem point for this anomaly.

[0058] As an optional implementation, the method may further include:

[0059] When the domain controller software is initially identified as having the aforementioned problem, a preset data backflow strategy is executed.

[0060] When the domain controller software generates incorrect control commands or fails to respond based on the data backfeed strategy, the problem point in the domain controller software is determined to be valid.

[0061] For example, during real-vehicle testing, if a vehicle exhibits an anomaly where it fails to change lanes correctly based on road conditions, the lane-change control module in the domain control software is initially identified as the source of the problem. In this case, a data feedback strategy is used to recreate the vehicle's autonomous driving parameters in a simulation environment detached from the hardware when the anomaly occurred. If the lane-change control module still exhibits an anomaly in the simulation environment, failing to generate correct lane-change control commands, then the lane-change control module is confirmed as the problem. If the lane-change control module generates correct lane-change control commands based on road conditions and controls the simulated vehicle to change lanes in the simulation environment, then the lane-change control module is determined to be the problem.

[0062] As an optional implementation method, executing a preset data backfeed strategy may include:

[0063] A vehicle simulation test environment is generated based on the real vehicle test data. The vehicle simulation test environment is used to characterize the normal autonomous driving conditions of the vehicle in the simulation test.

[0064] The domain control software is run in the vehicle simulation test environment;

[0065] The domain control software executes a corresponding test script based on the problem point. The test script carries environmental parameters, which are used to characterize the road environment features when the problem point occurs.

[0066] In this embodiment, the vehicle simulation test environment is used to characterize the vehicle's normal autonomous driving conditions during simulation testing. This can be understood as the vehicle simulation test environment converting the vehicle's autonomous driving conditions into vehicle-related visual data. This visual data can include road condition signals, vehicle longitudinal and lateral states, and the vehicle's autonomous driving planned route, etc.

[0067] In this embodiment, the test script can be understood as an environmental condition that triggers a problem. For example, during a real-vehicle test, when the vehicle's camera detects that the left lane is clear and the vehicle's radar speedometer detects that the vehicle in front is moving slowly, the camera and radar speedometer send parameter signals representing the detection results to the domain control software. The lane-change control module in the domain control software then generates corresponding control commands to control the vehicle to change lanes to the left. When the problem is initially identified as the lane-change control module, the domain control software is run in a simulation test environment, and the test script simulates the parameter signals sent by the camera and radar speedometer to the domain control software, serving as the environmental parameters that trigger the problem, to test the effectiveness of the lane-change control module as a problem under the same road conditions.

[0068] As an optional implementation, before determining the problem points in the domain control software based on the actual vehicle test data, the method may further include:

[0069] The actual vehicle test data is backed up as a .dat file.

[0070] By dividing real-vehicle test data into DAT files, it becomes easier to reuse the data multiple times, reducing the number of real-vehicle tests required for various tests, thereby lowering costs and improving safety. Simultaneously, saving and overlaying historical real-vehicle test data enriches the test scenarios in subsequent simulation tests, improving the efficiency of scenario transformation.

[0071] As an optional implementation, the method may further include:

[0072] The domain controller software will be iteratively updated based on the identified problems.

[0073] Based on the updated domain controller software, a preset regression verification strategy is executed;

[0074] When the domain controller software generates the correct control command based on the regression verification strategy, the repair result of the problem point is determined to be effective.

[0075] In this embodiment, the domain controller software developers iteratively update the domain controller software based on the issues identified in the test results. After the domain controller software is updated, testers use regression verification strategies to test the updated domain controller software to determine if the issues present before the update have been resolved.

[0076] For example, if the lane-change control module in the domain controller software is identified as a problem area, the domain controller software developers will update and fix it. After the update, testers will conduct autonomous driving simulation tests using regression verification strategies to recreate the scenario that triggered the problem. If, when a lane-change scenario is triggered, the lane-change control module generates the correct control command and controls the simulated vehicle to change lanes, the lane-change control module is considered to have been fixed and updated, meaning the fix for the domain problem is effective. If, when a lane-change scenario is triggered, the lane-change control module still generates incorrect control commands or remains unresponsive, the lane-change control module is considered not to have been fixed and updated, meaning the fix for the domain problem is invalid.

[0077] As an optional implementation, executing a preset regression verification strategy may include:

[0078] A vehicle simulation test environment is generated based on the actual vehicle test data.

[0079] The updated domain control software is run in the vehicle simulation test environment, wherein the updated domain control software is used to execute corresponding test scripts based on the problem points.

[0080] The vehicle simulation test environment can be understood as the simulated vehicle being in a normal autonomous driving state, and the test script can be understood as the environmental conditions that trigger the problem points.

[0081] In this embodiment, real-vehicle test data is transformed into visualized data to characterize the autonomous driving conditions of the simulated vehicle. This visualized data can include road condition signals, vehicle longitudinal and lateral states, and the vehicle's autonomous driving planned route, allowing testers to understand parameter changes during the execution of regression verification strategies. When the domain control software runs in the simulation test environment, test scripts generate environmental conditions that trigger problem points (such as the braking control module in the domain control software) (e.g., the simulated vehicle reaching the end point of the autonomous driving planned route). When the braking control module receives the end-point signal indicating the simulated vehicle has reached the end point of the autonomous driving planned route, it can generate the correct braking control command based on the signal and control the simulated vehicle to brake and stop. In this case, the repair result for the braking control module in the domain control software is deemed effective. Conversely, if the braking control module does not respond when it receives the end-point signal and the simulated vehicle remains in autonomous driving mode, the repair result for the braking control module in the domain control software is deemed invalid. This verifies whether the original problem points in the updated domain control software have been effectively repaired.

[0082] As an optional implementation, determining whether the real vehicle test data is abnormal based on a preset detection strategy may include:

[0083] When the front-end device senses that the lane environment meets the preset vehicle lane-changing conditions, and the domain control software cannot control the vehicle to perform the corresponding lane-changing operation according to the vehicle lane-changing conditions, it is determined that the real vehicle test data is abnormal.

[0084] When the front-end device senses that the distance between the vehicle and the solid lines on both sides of the lane is not equal or the vehicle deviates from the preset planned route, and the domain control software cannot control the vehicle to return to the correct driving trajectory, it determines that there is an anomaly in the real vehicle test data.

[0085] In this embodiment, when the front-end device of the vehicle undergoing real-vehicle testing detects that the vehicle in front is moving slowly and the left lane is clear, the vehicle needs to change lanes to the left. If the domain control software cannot generate a lane-changing control command and control the vehicle to change lanes to the left, it is determined that there is an anomaly in the real-vehicle test data.

[0086] Alternatively, when the front-end device of a vehicle undergoing real-vehicle testing detects that the vehicle deviates from the preset autonomous driving planning route, the vehicle needs to re-determine a route between the current position and the destination. If the domain control software cannot generate a control command to reset the planned route and control the vehicle to go to the destination based on the new planned route, then it is determined that the real-vehicle test data is abnormal.

[0087] Alternatively, if the front-end equipment of a vehicle undergoing real-vehicle testing detects that the vehicle is continuously obstructing the lane traffic lines during autonomous driving, the vehicle needs to move to the center of the lane. If the domain control software cannot generate a centering control command and control the vehicle to drive in the center of the lane, then it is determined that there is an anomaly in the real-vehicle test data.

[0088] Using the above method, real vehicle test data is collected through front-end devices deployed on the vehicle and backed up as a .dat file. Then, anomalies in the real vehicle test data are judged through a preset detection strategy.

[0089] When the front-end equipment of a vehicle undergoing real-vehicle testing detects that the vehicle in front is moving slowly and the left lane is clear, the vehicle needs to change lanes to the left. If the domain control software cannot generate a lane-changing control command and control the vehicle to change lanes to the left, it is determined that there is an anomaly in the real-vehicle test data.

[0090] Alternatively, when the front-end device of a vehicle undergoing real-vehicle testing detects that the vehicle deviates from the preset autonomous driving planning route, the vehicle needs to re-determine a route between the current position and the destination. If the domain control software cannot generate a control command to reset the planned route and control the vehicle to go to the destination based on the new planned route, then it is determined that the real-vehicle test data is abnormal.

[0091] Alternatively, if the front-end equipment of a vehicle undergoing real-vehicle testing detects that the vehicle is continuously obstructing the lane traffic lines during autonomous driving, the vehicle needs to move to the center of the lane. If the domain control software cannot generate a centering control command and control the vehicle to drive in the center of the lane, then it is determined that there is an anomaly in the real-vehicle test data.

[0092] After confirming the anomalies in the real vehicle test data, the corresponding control modules in the domain control software were identified as the problem points causing the anomalies, and the effectiveness of these problem points was tested. A vehicle simulation test environment was generated based on the real vehicle test data, and the domain control software was run within this environment. The domain control software then executed corresponding test scripts based on the problem points, triggering the environmental conditions that caused the aforementioned problem points in the simulated vehicle.

[0093] If the domain control software generates an incorrect control command or fails to respond when the problem point is triggered, then the problem point in the domain control software is determined to be valid; if the domain control software generates a correct control command and controls the simulated vehicle to perform the corresponding operation when the problem point is triggered, then the problem point in the domain control software is determined to be invalid.

[0094] When the issue is valid, it is fed back to the domain controller software developers as a test result, so that the developers can update and iterate the domain controller software based on the issue.

[0095] When the domain control software update is completed, the preset regression verification strategy is executed according to the new version of the domain control software. That is, the vehicle simulation test environment is generated again based on the real vehicle test data, and the new version of the domain control software is run in the vehicle simulation test environment. The new version of the domain control software executes the corresponding test scripts according to the problem points, and the test scripts trigger the environmental conditions where the problem points occur in the simulated vehicle.

[0096] If the new version of the domain controller software generates the correct control command and controls the simulated vehicle to perform the corresponding operation when the problem point is triggered, then the repair result of the problem point is determined to be effective; if the new version of the domain controller software still generates incorrect control commands or has no response when the problem point is triggered, then the repair result of the problem point is determined to be ineffective, and the problem point will be fed back to the domain controller software developers for repair until the repair result is effective.

[0097] Based on the above design, the testing of vehicle autonomous driving functions can be carried out without hardware, avoiding multiple real vehicle tests due to the complexity of autonomous driving testing requirements, reducing the cost of test benches, and simplifying the testing process.

[0098] Please refer to Figure 3 This application also provides an autonomous driving test device 200, which includes at least one software function module that can be stored in memory 102 in the form of software or firmware or embedded in the operating system (OS) of electronic device 100. Processor 101 is used to execute executable modules stored in memory 102, such as the software function modules and computer programs included in autonomous driving test device 200.

[0099] The autonomous driving test device 200 includes a data acquisition unit 210, a judgment unit 220, a first determination unit 230, a first execution unit, a second determination unit, a backup unit, an update unit, and a third determination unit. The functions of each unit are as follows:

[0100] The acquisition unit 210 is used to acquire real vehicle test data of the vehicle through the front-end device. The real vehicle test data includes a set of parameter signals used to characterize the autonomous driving conditions of the vehicle in the real vehicle test.

[0101] The judgment unit 220 is used to determine whether there is any abnormality in the actual vehicle test data based on a preset detection strategy;

[0102] The first determining unit 230 is used to determine the problem points in the domain control software based on the actual vehicle test data when there are abnormalities in the actual vehicle test data, and to take the problem points as test results. The problem points include the control modules corresponding to the parameter signals that caused the abnormalities.

[0103] Optionally, the autonomous driving test device 200 may also include:

[0104] The first execution unit is used to execute a preset data backflow strategy when the domain controller software is initially determined to have the problem point;

[0105] The second determining unit is used to determine that the problem point existing in the domain controller software is valid when the domain controller software generates an incorrect control command or has no response based on the data backfeed strategy.

[0106] Optionally, the first execution unit is further configured to:

[0107] A vehicle simulation test environment is generated based on the real vehicle test data. The vehicle simulation test environment is used to characterize the normal autonomous driving conditions of the vehicle in the simulation test.

[0108] The domain control software is run in the vehicle simulation test environment;

[0109] The domain control software executes a corresponding test script based on the problem point. The test script carries environmental parameters, which are used to characterize the road environment features when the problem point occurs.

[0110] Optionally, the autonomous driving test device 200 may also include:

[0111] The backup unit is used to back up the actual vehicle test data into a .dat file.

[0112] Optionally, the autonomous driving test device 200 may also include:

[0113] The update unit is used to iteratively update the domain controller software based on the problem points;

[0114] The second execution unit is used to execute a preset regression verification strategy based on the updated domain controller software;

[0115] The third determining unit is used to determine that the repair result of the problem point is valid when the domain control software generates the correct control command based on the regression verification strategy.

[0116] Optionally, the second execution unit is also used for:

[0117] A vehicle simulation test environment is generated based on the actual vehicle test data.

[0118] The updated domain control software is run in the vehicle simulation test environment, wherein the updated domain control software is used to execute corresponding test scripts based on the problem points.

[0119] Optionally, the determination unit 220 is also used for:

[0120] When the front-end device senses that the lane environment meets the preset vehicle lane-changing conditions, and the domain control software cannot control the vehicle to perform the corresponding lane-changing operation according to the vehicle lane-changing conditions, it is determined that the real vehicle test data is abnormal.

[0121] When the front-end device senses that the distance between the vehicle and the solid lines on both sides of the lane is not equal, and the domain control software cannot control the vehicle to return to the center position of the lane, it is determined that there is an anomaly in the real vehicle test data.

[0122] When the front-end device senses that the vehicle deviates from the preset planned route, and the domain control software is unable to replan the route between the current vehicle position and the destination, it determines that the real vehicle test data is abnormal.

[0123] In this embodiment, the processor 101 can be an integrated circuit chip with signal processing capabilities. The processor 101 can be a general-purpose processor. For example, it can be a central processing unit (CPU), 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, capable of implementing or executing the methods, steps, and logic block diagrams disclosed in the embodiments of this application.

[0124] The memory 102 can be, but is not limited to, random access memory, read-only memory, programmable read-only memory, erasable programmable read-only memory, electrically erasable programmable read-only memory, etc. In this embodiment, the memory 102 can be used to store actual vehicle test data, .dat files, etc. Of course, the memory 102 can also be used to store programs, which the processor 101 executes after receiving an execution instruction.

[0125] Understandable, Figure 1 The electronic device 100 shown is only a schematic diagram; the electronic device 100 may also include components that are more... Figure 1 More components are shown. Figure 1 The components shown can be implemented using hardware, software, or a combination thereof.

[0126] It should be noted that those skilled in the art will understand that, for the sake of convenience and brevity, the specific working process of the electronic device 100 described above can be referred to the corresponding process of each step in the aforementioned method, and will not be elaborated further here.

[0127] This application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program that, when run on a computer, causes the computer to perform the autonomous driving test method as described in the above embodiments.

[0128] Based on the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented by hardware or by using software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solution of this application can be embodied in the form of a software product. This software product can be stored in a non-volatile storage medium (such as CD-ROM, USB flash drive, mobile hard drive, etc.) and includes several instructions to cause a computer device (such as a personal computer, server, or network device, etc.) to execute the methods described in the various implementation scenarios of this application.

[0129] In summary, this application provides an autonomous driving testing method, apparatus, electronic device, and storage medium. In this solution, when anomalies are found in the real-vehicle test data, the corresponding control module in the domain control software is identified as the problem point causing the anomaly, and this problem point is fed back as a test result to the domain control software developers for repair. Thus, testing of vehicle autonomous driving functions can be achieved without hardware limitations, avoiding multiple real-vehicle tests due to the complexity of autonomous driving testing requirements, reducing bench costs for testing, and simplifying testing procedures.

[0130] In the embodiments provided in this application, it should be understood that the disclosed apparatus, systems, and methods can also be implemented in other ways. The apparatus, systems, and methods embodiments described above are merely illustrative. For example, the flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code, which includes one or more executable instructions for implementing a specified logical function. It should also be noted that each block in a block diagram and / or flowchart, and combinations of blocks in block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions. Furthermore, the functional modules in the various embodiments of this application can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.

[0131] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. An autonomous driving testing method, characterized in that, The method includes: Real-vehicle test data is collected through front-end devices. The real-vehicle test data includes a set of parameter signals used to characterize the autonomous driving conditions of the vehicle in the real-vehicle test. Based on a preset detection strategy, determine whether there are any anomalies in the actual vehicle test data; When there are abnormalities in the actual vehicle test data, the problem points in the domain control software are determined based on the actual vehicle test data, and the problem points are used as test results. The problem points include the control modules corresponding to the parameter signals that caused the abnormality. The method further includes: When the domain controller software is initially identified as having the problem, a preset data backflow strategy is executed. When the domain controller software generates an incorrect control command or fails to respond based on the data feedback strategy, it is determined that the problem point in the domain controller software is valid. Execute the preset data backfeed strategy, including: A vehicle simulation test environment is generated based on the real vehicle test data. The vehicle simulation test environment is used to characterize the normal autonomous driving conditions of the vehicle in the simulation test. The domain control software is run in the vehicle simulation test environment; The domain control software executes a corresponding test script based on the problem point. The test script carries environmental parameters, which are used to characterize the road environment features when the problem point occurs.

2. The method according to claim 1, characterized in that, Before determining the problem points in the domain control software based on the actual vehicle test data, the method further includes: The actual vehicle test data is backed up as a .dat file.

3. The method according to claim 1, characterized in that, The method further includes: The domain controller software will be iteratively updated based on the identified problems. Based on the updated domain controller software, a preset regression verification strategy is executed; When the domain controller software generates the correct control command based on the regression verification strategy, the repair result of the problem point is determined to be effective.

4. The method according to claim 3, characterized in that, Execute the preset regression validation strategy, including: A vehicle simulation test environment is generated based on the actual vehicle test data. The updated domain control software is run in the vehicle simulation test environment, wherein the updated domain control software is used to execute corresponding test scripts based on the problem points.

5. The method according to claim 1, characterized in that, Based on a preset detection strategy, determine whether the real vehicle test data contains anomalies, including: When the front-end device senses that the lane environment meets the preset vehicle lane-changing conditions, and the domain control software cannot control the vehicle to perform the corresponding lane-changing operation according to the vehicle lane-changing conditions, it is determined that the real vehicle test data is abnormal. When the front-end device senses that the distance between the vehicle and the solid lines on both sides of the lane is not equal, and the domain control software cannot control the vehicle to return to the center position of the lane, it is determined that there is an anomaly in the real vehicle test data. When the front-end device senses that the vehicle deviates from the preset planned route, and the domain control software is unable to replan the route between the current vehicle position and the destination, it is determined that the real vehicle test data is abnormal.

6. An autonomous driving testing device, characterized in that, The device includes The acquisition unit is used to acquire real vehicle test data of the vehicle through front-end equipment. The real vehicle test data includes a set of parameter signals used to characterize the autonomous driving conditions of the vehicle in the real vehicle test. The judgment unit is used to determine whether there are any abnormalities in the actual vehicle test data based on a preset detection strategy; The determining unit is used to determine the problem points in the domain control software based on the actual vehicle test data when there are abnormalities in the actual vehicle test data, and to take the problem points as test results. The problem points include the control modules corresponding to the parameter signals that caused the abnormalities. The determination unit is also used for: When the domain controller software is initially identified as having the problem, a preset data backflow strategy is executed. When the domain controller software generates an incorrect control command or fails to respond based on the data feedback strategy, it is determined that the problem point in the domain controller software is valid. Execute the preset data backfeed strategy, including: A vehicle simulation test environment is generated based on the real vehicle test data. The vehicle simulation test environment is used to characterize the normal autonomous driving conditions of the vehicle in the simulation test. The domain control software is run in the vehicle simulation test environment; The domain control software executes a corresponding test script based on the problem point. The test script carries environmental parameters, which are used to characterize the road environment features when the problem point occurs.

7. An electronic device, characterized in that, The electronic device includes a processor and a memory coupled together, the memory storing a computer program that, when executed by the processor, causes the electronic device to perform the method as described in any one of claims 1-5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when run on a computer, causes the computer to perform the method as described in any one of claims 1-5.

Images