Automatic driving simulation test case generalization method and system

Abstract

The invention provides an automatic driving simulation test case generalization method and system, and relates to the technical field of vehicle automatic driving test. Comprising the following steps: determining an attribute category corresponding to each function application scene test, and determining a plurality of numerical value intervals corresponding to each attribute category; performing permutation and combination on all the numerical value taking intervals of the attribute category corresponding to each function application scene to obtain various generalization test case parameters; taking each function application scene as an index, determining a static environment and a dynamic scene, embedding various generalization test case parameters corresponding to the function application scenes, and establishing a generalization test case database; and based on the simulation test target, determining a function application scene corresponding to the simulation test target, retrieving a generalization test case corresponding to the function application scene from the generalization test case database, and performing automatic driving simulation. According to the method, the generalization test case database can be constructed, the operation is simple and convenient, and the generalization effect is good.

Description

Technical Field

[0001] The present invention belongs to the technical field of vehicle autonomous driving testing, and in particular to a method and system for generalizing autonomous driving simulation test cases. Background Art

[0002] With the rapid development of autonomous driving technology, how to ensure the safety and reliability of autonomous driving systems has become a research focus. Traditional actual road testing methods are not only time-consuming and labor-intensive, but also fail to cover all possible scenarios. Therefore, verifying autonomous driving algorithms through simulation testing has become an effective solution.

[0003] Autonomous driving simulation testing provides a reliable and efficient means for the development and testing of autonomous driving algorithms by simulating various road conditions and traffic conditions in a virtual environment. It can cover a large number of test cases in a short period of time, evaluate system performance, correct defects and optimize algorithms, thereby greatly improving the safety and reliability of autonomous driving systems. Autonomous driving simulation testing includes test methods such as software-in-the-loop (SIL), model-in-the-loop (MIL), and hardware-in-the-loop (HIL), but without exception, the core test content needs to rely on test cases to achieve. Combined with the simulation scenario library and scenario generalization capabilities, on this basis, the software and model-in-the-loop method is adopted to test the functional logic of the autonomous driving algorithm, and the algorithm verification and optimization are completed by designing test cases.

[0004] Although autonomous driving simulation testing plays a vital role in the research and development of autonomous driving systems, and can significantly shorten the development cycle, improve testing efficiency and reduce the risks of actual road testing, current simulation testing methods often have problems such as insufficient test case coverage, low testing efficiency and poor generalization ability.

[0005] Autonomous driving systems need to cope with a variety of complex road conditions and traffic conditions, including different weather conditions, different road conditions, and different behaviors of traffic participants. However, due to limited test scenario resources, current simulation test methods often rely too much on specific test cases and scenarios, lack sufficient generalization capabilities to cope with unknown challenges, and testers find it difficult to design and implement test cases that cover all possible scenarios, which may cause some potential defects and problems to be missed during the testing phase. In actual applications, the system cannot adapt well to complex and changing environments and scenarios, which reduces the reliability and safety of the system.

[0006] On the other hand, due to the complexity and diversity of autonomous driving systems, testers need to spend a lot of time and energy to design and execute test cases. In addition, the testing process requires multiple rounds of iteration and optimization to ensure the performance and safety of the system. This cumbersome testing process is not only time-consuming and labor-intensive, but may also reduce testing efficiency and affect project progress and delivery time. Summary of the invention

[0007] In order to overcome the deficiencies of the above-mentioned prior art, the present invention provides a method and system for generalizing autonomous driving simulation test cases. By determining a variety of dangerous function application scenarios and the attribute categories corresponding to the test dangerous function application scenarios, and the permutations and combinations of the attribute category numerical value ranges, a generalized test case parameter set corresponding to the above-mentioned dangerous function application scenarios is obtained, and then a generalized test case database is constructed. It only needs to index the function application scenarios to obtain a variety of generalized test cases. The operation is simple and the generalization effect is good.

[0008] To achieve the above objectives, one or more embodiments of the present invention provide the following technical solutions:

[0009] A first aspect of the present invention provides a method for generalizing autonomous driving simulation test cases.

[0010] The generalization method of autonomous driving simulation test cases includes the following steps:

[0011] Construct multiple functional application scenarios for autonomous driving simulation testing;

[0012] Determine the attribute category corresponding to each functional application scenario test, and determine the multiple value ranges corresponding to each attribute category;

[0013] All numerical value intervals of the attribute categories corresponding to each functional application scenario are freely arranged and combined to obtain a variety of generalized test case parameters, thereby forming a generalized test case parameter set corresponding to the functional application scenario;

[0014] Taking each functional application scenario as an index, determine the static environment and dynamic scenario, and embed multiple generalized test case parameters in the generalized test case parameter set corresponding to the functional application scenario to establish a generalized test case database;

[0015] Based on the simulation test target, the functional application scenario corresponding to the simulation test target is determined, the generalized test case corresponding to the functional application scenario is retrieved from the generalized test case database, and the autonomous driving simulation is performed based on the generalized test case.

[0016] As an optional technical solution, the various functional application scenarios of the autonomous driving simulation test include dangerous functional application scenarios, including the host car avoidance scenario when the target vehicle falls off an obstacle, the host car avoidance scenario when pedestrians cross the road, and the host car avoidance scenario when the target vehicle forcibly changes lanes.

[0017] As an optional technical solution, in the target vehicle falling obstacle and the host vehicle avoiding scene, the attribute categories include the target vehicle speed, the host vehicle speed, the lanes where the host vehicle and the target vehicle are located, whether the host vehicle is turning, the three-dimensional coordinates of the obstacle falling moment, the host vehicle coordinates at the moment the obstacle falls, the wind speed, the moment when the host vehicle detects the obstacle falling, and the braking deceleration of the host vehicle;

[0018] In the pedestrian road-crossing host vehicle avoidance scenario, the attribute categories include the host vehicle speed, whether the host vehicle is turning, the pedestrian's speed, the pedestrian's initial coordinate position, the host vehicle's coordinate position when the pedestrian starts to move, the time when the host vehicle detects the pedestrian, and the host vehicle's braking deceleration;

[0019] In the target vehicle forced lane change and host vehicle avoidance scenario, the attribute categories include the target vehicle speed, the host vehicle speed, whether the host vehicle is turning, the target vehicle lane change coordinate position, the host vehicle coordinate position at the time of the target vehicle lane change, the target vehicle lane change time, the time when the host vehicle detects the target vehicle, and the host vehicle braking deceleration.

[0020] As an optional technical solution, the multiple generalized test case parameters cover all attribute categories of corresponding functional application scenarios and all numerical value ranges of the attribute categories.

[0021] As an optional technical solution, each functional application scenario is used as an index to determine the static environment and dynamic scenario, and at the same time, multiple generalized test case parameters in the generalized test case parameter set corresponding to the functional application scenario are embedded to establish a generalized test case database, which specifically includes:

[0022] Determine the static environment, including determining the specific structure of the test road built in the simulation test software;

[0023] Determining a dynamic scenario, including determining a dynamic traffic flow set in the static environment;

[0024] Based on a specific functional application scenario, an object to be simulated is determined in the dynamic traffic flow, and generalized test case parameters are applied to the determined simulation object to construct a plurality of generalized test cases;

[0025] Based on multiple generalized test cases, a generalized test case database is established.

[0026] As optional technical solutions, it also includes:

[0027] Create test configuration engineering files based on the automated test platform, and deploy the autonomous driving algorithm model, dynamics model, and simulation test environment to the automated test platform;

[0028] Build and execute specific test steps for test cases on the automated testing platform;

[0029] Obtain the signals required for evaluation indicators in specific test steps, perform data analysis, generate a test report based on the data analysis results, and complete the development of test cases.

[0030] As optional technical solutions, it also includes:

[0031] After creating the test configuration project file, run it to ensure the connection between the automated test platform and the autonomous driving algorithm model, dynamics model, and simulation test environment;

[0032] If the link is unsuccessful, modify the test configuration project file according to the error.

[0033] A second aspect of the present invention provides an autonomous driving simulation test case generalization system.

[0034] The generalization system of autonomous driving simulation test cases includes:

[0035] The functional application scenario building module is configured to: build multiple functional application scenarios for autonomous driving simulation testing;

[0036] The attribute determination module is configured to: determine the attribute category corresponding to each functional application scenario test, and determine multiple numerical value ranges corresponding to each attribute category;

[0037] The parameter determination module is configured to: freely arrange and combine all numerical value intervals of the attribute category corresponding to each functional application scenario to obtain multiple generalized test case parameters, thereby forming a generalized test case parameter set corresponding to the functional application scenario;

[0038] The test case establishment module is configured to: determine the static environment and the dynamic scenario by taking each functional application scenario as an index, and embed a plurality of generalized test case parameters in the generalized test case parameter set corresponding to the functional application scenario to establish a generalized test case database;

[0039] The retrieval module is configured to: determine the functional application scenario corresponding to the simulation test target based on the simulation test target, retrieve the generalized test case corresponding to the functional application scenario from the generalized test case database, and perform autonomous driving simulation based on the generalized test case.

[0040] A third aspect of the present invention provides a computer-readable storage medium having a program stored thereon, which, when executed by a processor, implements the steps in the method for generalizing autonomous driving simulation test cases as described in the first aspect of the present invention.

[0041] The fourth aspect of the present invention provides an electronic device, including a memory, a processor, and a program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps in the method for generalizing autonomous driving simulation test cases as described in the first aspect of the present invention are implemented.

[0042] One or more of the above technical solutions have the following beneficial effects:

[0043] The present invention provides a method and system for generalizing test cases for autonomous driving simulation, aiming to improve the generalization, coverage, efficiency and reliability of autonomous driving system testing. By determining a variety of dangerous function application scenarios and the attribute categories corresponding to the test dangerous function application scenarios, and the permutations and combinations of the attribute category numerical value ranges, a generalized test case parameter set corresponding to the above-mentioned dangerous function application scenarios is obtained, and then a generalized test case database is constructed. It only needs to index the function application scenarios to obtain a variety of generalized test cases, which is easy to operate and has a good generalization effect.

[0044] The various functional application scenarios of the autonomous driving simulation test provided by the present invention include a scenario in which the target vehicle drops an obstacle and the host vehicle avoids it, a scenario in which pedestrians cross the road and the host vehicle avoids it, and a scenario in which the target vehicle forcibly changes lanes and the host vehicle avoids it. Attribute categories are designed respectively for the above-mentioned dangerous functional application scenarios. Based on the attribute categories and the numerical value ranges of the attribute categories, generalized test cases can be generated taking into account the dangerous functional application scenarios.

[0045] The present invention is based on a simulation scenario library of tens of thousands of cases, and uses automated testing software to quickly build a test case template that includes test steps, signal extraction, indicator evaluation, and automated report generation, so as to solve the drawbacks of insufficient coverage and manual construction of current simulation test cases; the provided autonomous driving simulation test case generalization method can solve the problem of low generalization of current test cases.

[0046] The standard test case template for autonomous driving simulation testing obtained based on the autonomous driving simulation test case development method is reusable and scalable, and automatically realizes the design of test cases and test output, effectively solving the time-consuming and labor-intensive drawbacks of traditional test case testing.

[0047] The autonomous driving simulation test case generalization method provided by the present invention integrates the acquired scene file name list with the key parameter generalization list, and based on the integration results, automatically generates generalized test cases in batches for autonomous driving function simulation testing, which can solve the current problem of low generalization of test cases.

[0048] Advantages of additional aspects of the present invention will be given in part in the following description, and in part will become obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] The accompanying drawings in the specification, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0050] Figure 1 This is a flowchart of a specific execution of the method of the first embodiment.

[0051] Figure 2 A flowchart is specifically generated for the generalized test case of the first embodiment. DETAILED DESCRIPTION

[0052] It should be noted that the following detailed descriptions are exemplary and are intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present invention belongs.

[0053] It should be noted that the terms used herein are for describing specific embodiments only and are not intended to be limiting of exemplary embodiments according to the present invention.

[0054] In the absence of conflict, the embodiments of the present invention and the features of the embodiments may be combined with each other.

[0055] Embodiment 1

[0056] As mentioned above, due to the limited resources of test scenarios, current simulation test methods often rely too much on specific test cases and scenarios, lack sufficient generalization capabilities to cope with unknown challenges, and it is difficult for testers to design and implement test cases that cover all possible scenarios, which may cause some potential defects and problems to be missed during the testing phase. In actual applications, the system cannot adapt well to complex and changing environments and scenarios, which reduces the reliability and security of the system.

[0057] Especially for some more dangerous scenarios, if there is a lack of simulation of these dangerous function application scenarios, or the simulation of dangerous function application scenarios lacks generalization and cannot cover all scenarios well, it will be detrimental to improving the reliability and safety of simulation tests.

[0058] On the other hand, due to the complexity and diversity of autonomous driving systems, testers need to spend a lot of time and energy to design and execute test cases. In addition, the testing process requires multiple rounds of iteration and optimization to ensure the performance and safety of the system. This cumbersome testing process is not only time-consuming and labor-intensive, but may also reduce testing efficiency and affect project progress and delivery time.

[0059] Based on this, this embodiment discloses a method for generalizing automatic driving simulation test cases. Figure 1As shown, the following steps may be included:

[0060] Construct multiple functional application scenarios for autonomous driving simulation testing;

[0061] Determine the attribute category corresponding to each functional application scenario test, and determine the multiple value ranges corresponding to each attribute category;

[0062] All numerical value intervals of the attribute categories corresponding to each functional application scenario are freely arranged and combined to obtain a variety of generalized test case parameters, thereby forming a generalized test case parameter set corresponding to the functional application scenario;

[0063] Taking each functional application scenario as an index, determine the static environment and dynamic scenario, and embed multiple generalized test case parameters in the generalized test case parameter set corresponding to the functional application scenario to establish a generalized test case database;

[0064] Based on the simulation test target, the functional application scenario corresponding to the simulation test target is determined, the generalized test case corresponding to the functional application scenario is retrieved from the generalized test case database, and the autonomous driving simulation is performed based on the generalized test case.

[0065] In some embodiments, the various functional application scenarios of the autonomous driving simulation test include dangerous functional application scenarios, including the host vehicle avoidance scenario when the target vehicle falls off an obstacle, the host vehicle avoidance scenario when pedestrians cross the road, and the host vehicle avoidance scenario when the target vehicle forcibly changes lanes.

[0066] In some embodiments, in the target vehicle falling obstacle and the host vehicle avoiding scenario, the attribute categories include the target vehicle speed, the host vehicle speed, the lanes where the host vehicle and the target vehicle are located, whether the host vehicle is turning, the three-dimensional coordinates of the obstacle falling moment, the host vehicle coordinates of the obstacle falling moment, the wind speed, the moment when the host vehicle detects the obstacle falling, and the braking deceleration of the host vehicle;

[0067] In the pedestrian road-crossing host vehicle avoidance scenario, the attribute categories include the host vehicle speed, whether the host vehicle is turning, the pedestrian's speed, the pedestrian's initial coordinate position, the host vehicle's coordinate position when the pedestrian starts to move, the time when the host vehicle detects the pedestrian, and the host vehicle's braking deceleration;

[0068] In the target vehicle forced lane change and host vehicle avoidance scenario, the attribute categories include the target vehicle speed, the host vehicle speed, whether the host vehicle is turning, the target vehicle lane change coordinate position, the host vehicle coordinate position at the time of the target vehicle lane change, the target vehicle lane change time, the time when the host vehicle detects the target vehicle, and the host vehicle braking deceleration.

[0069] In some embodiments, the multiple generalized test case parameters cover all attribute categories of corresponding functional application scenarios and all numerical value ranges of the attribute categories.

[0070] This embodiment can effectively deal with the omission of the above-mentioned scenarios in the prior art by classifying the above-mentioned dangerous function application scenarios, and cover all possible dangerous scenarios as much as possible. And by dividing the attribute categories of each dangerous function application scenario and arranging and combining the value ranges of the attribute categories, as many generalized test case parameters as possible with comprehensive coverage can be obtained, and then multiple generalized test cases can be obtained, which improves the rationality of the autonomous driving simulation test cases.

[0071] As a further approach:

[0072] Taking the scenario where the target vehicle drops an obstacle and the host vehicle avoids it as an example, in matching the numerical value ranges of the above attribute categories, the vehicle size information can be matched by the target vehicle and the host vehicle type, and the obstacle type, size, color, and three-dimensional coordinate information at the time of dropping, the host vehicle position at the time of obstacle dropping, and the host vehicle speed can be matched with its vehicle size information, thereby achieving more efficient parameter set matching.

[0073] In some embodiments, each functional application scenario is used as an index to determine the static environment and the dynamic scenario, and a plurality of generalized test case parameters in the generalized test case parameter set corresponding to the functional application scenario are embedded to establish a generalized test case database, specifically including:

[0074] Determine the static environment, including determining the specific structure of the test road built in the simulation test software;

[0075] Determining a dynamic scenario, including determining a dynamic traffic flow set in the static environment;

[0076] Based on a specific functional application scenario, an object to be simulated is determined in the dynamic traffic flow, and generalized test case parameters are applied to the determined simulation object to construct a plurality of generalized test cases;

[0077] Based on multiple generalized test cases, a generalized test case database is established.

[0078] Furthermore, in the above process:

[0079] The static environment specifically includes lane type information, lane quantity information, lane width information, lane line type information, traffic sign information and / or information board information;

[0080] Traffic participant information in a dynamic traffic flow is set, wherein the traffic participant information includes initial state information of a host vehicle and a target vehicle / pedestrian, vehicle type information, and / or triggered vehicle / pedestrian behavior information.

[0081] In some embodiments, it also includes:

[0082] Create test configuration engineering files based on the automated test platform, and deploy the autonomous driving algorithm model, dynamics model, and simulation test environment to the automated test platform;

[0083] Build and execute specific test steps for test cases on the automated testing platform;

[0084] Obtain the signals required for evaluation indicators in specific test steps, perform data analysis, generate a test report based on the data analysis results, and complete the development of test cases.

[0085] In some embodiments, it also includes:

[0086] After creating the test configuration project file, run it to ensure the connection between the automated test platform and the autonomous driving algorithm model, dynamics model, and simulation test environment;

[0087] If the link is unsuccessful, modify the test configuration project file according to the error.

[0088] When specifically implementing the above autonomous driving simulation test case generalization method:

[0089] Based on the simulation test target, determine the functional application scenario corresponding to the simulation test target;

[0090] Extracting the function application scenario file names corresponding to the simulation test targets from the generalized test case database to form a scenario file name list;

[0091] Extract the generalized test case parameter set corresponding to the functional application scenario to form a generalized list of key parameters;

[0092] Integrate the scene file name list with the generalized list of key parameters;

[0093] Based on the integration results, generalized test cases are automatically generated in batches for simulation testing of autonomous driving functions.

[0094] When combining the list of scene file names with the generalized list of key parameters:

[0095] The generalized test cases corresponding to each scenario file name are extracted, and all the extracted generalized test cases are integrated into the integrated results.

[0096] Furthermore, combined with Figure 1 , the above specific execution process of this embodiment includes:

[0097] 1. Analyze the functional application scenarios according to the simulation test objectives. For example, when testing the high-speed navigation driving assistance H-NOP function, the scenarios that meet its high-speed and navigation characteristics should be selected;

[0098] 2. Extract scenarios for simulation testing based on Chery's virtual simulation test scenario library;

[0099] 3. Use the self-developed script tool ET_XOSC2.py to extract the scene file names for simulation testing and form a scene file name list;

[0100] 4. Use the self-developed script tool Get_testcase_rev0.py to extract the key parameters of the test case template built on the automated testing platform (such as scenario name, set initial vehicle speed, Trigger trigger waiting time, running time, etc.), enrich the test case automation generalization dimension, and form a generalized list of key parameters;

[0101] 5. Integrate the extracted simulation scenario file name list and the extracted test case template key parameters into an Excel table;

[0102] 6. Finally, for the Excel table integrated in step 5, execute the self-developed Python script tool to automatically generate generalized test cases in batches for autonomous driving function simulation testing.

[0103] Next, the specific method for generating generalized test cases in this embodiment will be further described in detail.

[0104] As mentioned above, current simulation testing methods often have problems such as insufficient test case coverage, low test efficiency, and weak generalization ability. It is difficult for testers to design and implement test cases that cover all possible scenarios. If they rely too much on specific test cases and scenarios, they will lack sufficient generalization ability to cope with unknown challenges, which may cause some potential defects and problems to be missed during the testing phase, thereby reducing the reliability and security of the system.

[0105] In addition, testers need to spend a lot of time and energy to design and execute test cases. The cumbersome testing process is not only time-consuming and labor-intensive, but may also reduce testing efficiency and affect project progress and delivery time.

[0106] Based on this, this embodiment also provides a specific method for generating a generalized test case for an autonomous driving simulation, including the following steps:

[0107] Determine the function test evaluation indicators and the signals required for the evaluation indicators based on the autonomous driving function test requirements;

[0108] Prepare the test case building environment, including software that provides simulation test environment and real vehicle dynamics model, autonomous driving algorithm model, functional DUT model and automated test platform;

[0109] Configure interfaces for software that provides simulation test environments and real vehicle dynamics models;

[0110] Create test configuration engineering files based on the automated test platform, and deploy the autonomous driving algorithm model, dynamics model, and simulation test environment to the automated test platform;

[0111] Build and execute specific test steps for test cases on the automated testing platform;

[0112] Obtain the signals required for evaluation indicators in specific test steps, perform data analysis, generate a test report based on the data analysis results, and complete the development of test cases.

[0113] The above process is based on a simulation scenario library of tens of thousands of cases. Through automated testing software, a test case template including test steps, signal extraction, indicator evaluation and automated report generation is quickly built to solve the current shortcomings of insufficient simulation test case coverage and manual construction. Specifically, the following steps may be included:

[0114] Step 1: First, according to the requirements of autonomous driving function testing, sort out the function test evaluation indicators according to standards, regulations and other documents, and analyze the signals required for sorting out the evaluation indicators;

[0115] Step 2: Prepare the test case environment, including VTD software that provides a simulation test environment, Carsim software required to provide a real vehicle dynamics model, an autonomous driving algorithm model, a functional DUT simulink model, and an automated test platform ECU-TEST;

[0116] Step 3: Configure the interfaces of each software platform, including commenting out the VTD built-in dynamic model and modifying the parameters in its parameters, and creating a new database in Carsim software to configure the required dynamic model;

[0117] Step 4: The automated test software ECU-TEST accesses VTD data and performs related operations through the SSH service, so the SSH service needs to be installed in advance to generate the RSA key;

[0118] Step 5: Create test configuration .tbc and .tcf project files based on ECU-TEST. By configuring the project files, deploy the autonomous driving function algorithm model, dynamics model and simulation test scenario environment to this automated test platform;

[0119] Step 6: Create a test case .pkg file based on the automated test platform ECU-TEST. The link between the automated test platform and the algorithm and simulation environment is established through the test configuration file in step 5, that is, the test steps can be built on the test platform;

[0120] Step 7: Signal recordings:

[0121] In the test step, all signals in the simulation process are recorded, and then all signals required for data analysis and indicator evaluation are pulled out in the Signal recordings window;

[0122] Step 8: Trace analysis data analysis:

[0123] After editing the test case steps and pulling all the required signals in the Signal recordings window, start Trace analysis data post-processing and indicator evaluation design and analysis;

[0124] Step 9: Generate a test report:

[0125] After editing the test steps, signal recording, and evaluation indicators, the entire test case is developed and built completely, and the test report can be automatically generated based on this automated testing platform.

[0126] The following will be combined with the attached Figure 2 , the specific implementation method of the above generation process in this embodiment is further described in detail.

[0127] 1. First, according to the requirements of autonomous driving function testing, sort out the function test evaluation indicators according to standards, regulations and other documents, and analyze the signals required for sorting out the evaluation indicators.

[0128] Taking the test of the ICA function of autonomous driving as an example, we first need to analyze and sort out the performance evaluation indicators of its function, such as speed control error, lateral deviation, lateral and longitudinal acceleration, following distance, TTC, THW, starting response time, etc., and then list the calculation methods or rules of each evaluation indicator to sort out the signals involved;

[0129] 2. Prepare the test case environment, including VTD software that provides a simulation test environment, Carsim software required to provide a real vehicle dynamics model, Simulink model of the autonomous driving algorithm function DUT, and ECU-TEST, an automated test platform;

[0130] 3. Configure the interfaces of each software platform, including commenting out the dynamic model that comes with VTD and modifying the parameters in its parameters: mainly modify the interface and portType in RDB and RDB_2, and create a new database in Carsim software to configure the required dynamic model;

[0131] 4. The automated test software ECU-TEST accesses VTD data and performs related operations through the SSH service. Therefore, it is necessary to install the SSH service in advance and enable the SSH service of the Linux host. Then enter the terminal and execute the command ssh-keygen-t rsa to generate the RSA key. Copy the id_rsa file in the .ssh folder and put it in the workspace\parameters folder of ECU-TEST in the Windows host for subsequent use.

[0132] 5. Create test configuration .tbc and .tcf project files based on ECU-TEST, and deploy the autonomous driving function algorithm model, dynamics model and simulation test scenario environment to this automated testing platform by configuring the project files.

[0133] After creating the .tbc and .tcf project files, run them to ensure the link between the automated test platform ECU-TEST and other tools. If the link fails, modify the configuration file according to the error message.

[0134] 6. Create a test case PKG. Through the test configuration file in step 5, a link is established between the automated test platform and the algorithm and simulation environment. That is, the test steps can be built on the test platform by pulling the tools in the toolbar.

[0135] In SIL / MIL testing (i.e., using Matlab / Simulink models as carriers), the loading scenario step of ECU-TEST requires VTD to feedback an RDB signal before the loading step is completed. However, in order for VTD to feedback the RDB signal, the simulation of the VTD scenario needs to be started. ECU-TEST has already started the simulation of VTD during loading, but VTD must start the simulation with Simulink before it can actually send the RDB to ECU-TEST.

[0136] Therefore, a .pkg file is required to run in parallel, start the simulink simulation while loading the VTD scene. Otherwise, the error "Timeout after 60s" will be reported;

[0137] 7. Signal recordings signal pull:

[0138] In the test step, all signals in the simulation process are recorded, and then all signals required for data analysis and indicator evaluation are pulled in the Signal recordings window. The main sources of signal categories are divided into two parts: one is the corresponding signal in the Plantmodel, that is, the Simulink model; the other is the RDB signal packaged and parsed under the VTD default port 48190;

[0139] 8. Trace analysis data analysis:

[0140] After editing the test case steps and pulling all the required signals in the Signal recordings window, start Trace analysis data post-processing. Here, the performance indicators of the autonomous driving function are evaluated through calculation formulas and signal processing.

[0141] 9. Generate test report:

[0142] After editing the test steps, signal recording, and evaluation indicators, the entire test case is developed and built completely. Based on this automated testing platform, the test report can be automatically generated. Then, the indicator design can be judged to be reasonable based on the indicator diagram in the test report, and the evaluation indicators can be optimized and iterated.

[0143] Based on the above steps, a standard test case template for autonomous driving simulation testing can be obtained. At the same time, this template is reusable and scalable, and can automatically implement the design of test cases and test output, effectively solving the time-consuming and labor-intensive drawbacks of traditional test case testing.

[0144] Embodiment 2

[0145] This embodiment discloses an autonomous driving simulation test case generalization system.

[0146] The generalization system of autonomous driving simulation test cases includes:

[0147] The functional application scenario building module is configured to: build multiple functional application scenarios for autonomous driving simulation testing;

[0148] The attribute determination module is configured to: determine the attribute category corresponding to each functional application scenario test, and determine multiple numerical value ranges corresponding to each attribute category;

[0149] The parameter determination module is configured to: freely arrange and combine all numerical value intervals of the attribute category corresponding to each functional application scenario to obtain multiple generalized test case parameters, thereby forming a generalized test case parameter set corresponding to the functional application scenario;

[0150] The test case establishment module is configured to: determine the static environment and the dynamic scenario by taking each functional application scenario as an index, and embed a plurality of generalized test case parameters in the generalized test case parameter set corresponding to the functional application scenario to establish a generalized test case database;

[0151] The retrieval module is configured to: determine the functional application scenario corresponding to the simulation test target based on the simulation test target, retrieve the generalized test case corresponding to the functional application scenario from the generalized test case database, and perform autonomous driving simulation based on the generalized test case.

[0152] Embodiment 3

[0153] The purpose of this embodiment is to provide a computer-readable storage medium.

[0154] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps in the method for generalizing autonomous driving simulation test cases as described in Example 1 of the present disclosure.

[0155] Embodiment 4

[0156] The purpose of this embodiment is to provide an electronic device.

[0157] An electronic device includes a memory, a processor, and a program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps in the method for generalizing autonomous driving simulation test cases as described in Example 1 of the present disclosure are implemented.

[0158] The steps involved in the apparatuses of the above embodiments 2, 3 and 4 correspond to the method embodiment 1, and the specific implementation methods can refer to the relevant description part of embodiment 1. The term "computer-readable storage medium" should be understood as a single medium or multiple media including one or more instruction sets; it should also be understood to include any medium that can store, encode or carry an instruction set for execution by a processor and enable the processor to execute any method in the present invention.

[0159] Those skilled in the art should understand that the modules or steps of the present invention described above can be implemented by a general-purpose computer device, or alternatively, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. The present invention is not limited to any specific combination of hardware and software.

[0160] Although the above describes the specific implementation mode of the present invention in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art on the basis of the technical solution of the present invention without creative work are still within the scope of protection of the present invention.

Claims

1. A generalization method for autonomous driving simulation test cases, characterized in that: The following steps are involved: Construct multiple functional application scenarios for autonomous driving simulation testing; Determine the attribute category corresponding to each functional application scenario test, and determine the multiple value ranges corresponding to each attribute category; All numerical value intervals of the attribute categories corresponding to each functional application scenario are freely arranged and combined to obtain a variety of generalized test case parameters, thereby forming a generalized test case parameter set corresponding to the functional application scenario; Taking each functional application scenario as an index, determine the static environment and dynamic scenario, and embed multiple generalized test case parameters in the generalized test case parameter set corresponding to the functional application scenario to establish a generalized test case database; Based on the simulation test target, the functional application scenario corresponding to the simulation test target is determined, the generalized test case corresponding to the functional application scenario is retrieved from the generalized test case database, and the autonomous driving simulation is performed based on the generalized test case.

2. The method for generalizing an autonomous driving simulation test case according to claim 1, characterized in that: The various functional application scenarios of the autonomous driving simulation test include dangerous functional application scenarios, including the host car avoidance scenario when the target vehicle falls off an obstacle, the host car avoidance scenario when pedestrians cross the road, and the host car avoidance scenario when the target vehicle forcibly changes lanes.

3. The method for generalizing an autonomous driving simulation test case according to claim 2, wherein: In the target vehicle falling obstacle and the host vehicle avoiding scenario, the attribute categories include the target vehicle speed, the host vehicle speed, the lanes where the host vehicle and the target vehicle are located, whether the host vehicle is turning, the three-dimensional coordinates of the obstacle falling moment, the host vehicle coordinates of the obstacle falling moment, the wind speed, the moment when the host vehicle detects the obstacle falling, and the braking deceleration of the host vehicle; In the pedestrian road-crossing host vehicle avoidance scenario, the attribute categories include the host vehicle speed, whether the host vehicle is turning, the pedestrian's speed, the pedestrian's initial coordinate position, the host vehicle's coordinate position when the pedestrian starts to move, the time when the host vehicle detects the pedestrian, and the host vehicle's braking deceleration; In the target vehicle forced lane change and host vehicle avoidance scenario, the attribute categories include the target vehicle speed, the host vehicle speed, whether the host vehicle is turning, the target vehicle lane change coordinate position, the host vehicle coordinate position at the time of the target vehicle lane change, the target vehicle lane change time, the time when the host vehicle detects the target vehicle, and the host vehicle braking deceleration.

4. The method for generalizing an autonomous driving simulation test case according to claim 1, wherein: The various generalized test case parameters cover all attribute categories of corresponding functional application scenarios and all numerical value ranges of the attribute categories.

5. The method for generalizing automatic driving simulation test cases according to claim 1, characterized in that: Taking each functional application scenario as an index, determine the static environment and dynamic scenario, and embed multiple generalized test case parameters in the generalized test case parameter set corresponding to the functional application scenario to establish a generalized test case database, including: Determine the static environment, including determining the specific structure of the test road built in the simulation test software; Determining a dynamic scenario, including determining a dynamic traffic flow set in the static environment; Based on a specific functional application scenario, an object to be simulated is determined in the dynamic traffic flow, and generalized test case parameters are applied to the determined simulation object to construct a plurality of generalized test cases; Based on multiple generalized test cases, a generalized test case database is established.

6. The method for generalizing automatic driving simulation test cases according to claim 1, characterized in that: Also includes: Create a test configuration engineering file based on the automated test platform, and deploy the autonomous driving algorithm model, dynamics model, and simulation test environment to the automated test platform; Build and execute specific test steps for test cases on the automated testing platform; Obtain the signals required for evaluation indicators in specific test steps, perform data analysis, generate a test report based on the data analysis results, and complete the development of test cases.

7. The method for generalizing an autonomous driving simulation test case according to claim 6, wherein: Also includes: After creating the test configuration project file, run it to ensure the connection between the automated test platform and the autonomous driving algorithm model, dynamics model, and simulation test environment; If the link is unsuccessful, modify the test configuration project file according to the error.

8. The automatic driving simulation test case generalization system is characterized by: include: The functional application scenario building module is configured to: build multiple functional application scenarios for autonomous driving simulation testing; The attribute determination module is configured to: determine the attribute category corresponding to each functional application scenario test, and determine multiple numerical value ranges corresponding to each attribute category; The parameter determination module is configured to: freely arrange and combine all numerical value intervals of the attribute category corresponding to each functional application scenario to obtain multiple generalized test case parameters, thereby forming a generalized test case parameter set corresponding to the functional application scenario; The test case establishment module is configured to: determine the static environment and the dynamic scenario by taking each functional application scenario as an index, and embed a plurality of generalized test case parameters in the generalized test case parameter set corresponding to the functional application scenario to establish a generalized test case database; The retrieval module is configured to: determine the functional application scenario corresponding to the simulation test target based on the simulation test target, retrieve the generalized test case corresponding to the functional application scenario from the generalized test case database, and perform autonomous driving simulation based on the generalized test case.

9. A computer-readable storage medium having a program stored thereon, characterized in that: When the program is executed by a processor, the steps in the method for generalizing an autonomous driving simulation test case as described in any one of claims 1 to 7 are implemented.

10. An electronic device comprising a memory, a processor, and a program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps in the autonomous driving simulation test case generalization method as described in any one of claims 1 to 7 are implemented.

Images