Automatic driving test scene automatic construction method and device, equipment and storage medium

Abstract

The present disclosure relates to an automatic driving test scene automatic construction method, device and equipment and storage medium, the method comprises the following steps: constructing a test scene, dynamically configuring the test scene to obtain a plurality of case scenes, and putting them into a training queue; the case scenes in the training queue are run in turn, the running result of the host vehicle in the current case scene is obtained, whether the running result exists a preset behavior is judged, in response to the existence of the preset behavior, the current case scene is recorded to a first scene library, in response to the non-existence of the preset behavior, the current case scene is recorded to a second scene library; the running result of the case scene in the second scene library is obtained, whether the running result produces a preset feedback behavior is judged, and based on the judgment result, the case scene corresponding to the judgment result is triggered to enhance the test or is deleted. The present disclosure can construct key scenes based on preset standard scenes, improve scene generation efficiency and quality, improve test effect, and reduce test cost.

Description

Technical Field

[0001] This disclosure relates to the field of autonomous driving testing technology, and in particular to a method, apparatus, device and storage medium for automatically constructing autonomous driving test scenarios. Background Technology

[0002] Autonomous driving has become a hot research area in recent years. However, due to the complex and ever-changing traffic scenarios and the high cost of testing, traditional open road testing and closed test track testing are difficult to meet the stringent requirements of autonomous driving technology for reliability and robustness. If the performance of autonomous vehicles in real-world scenarios is tested, the cost is high and may cause actual losses. Therefore, there is an urgent need for a method to test the performance of autonomous vehicles in such scenarios through simulation. Simulation based on digital virtual simulation technology has become one of the important means of autonomous driving simulation testing.

[0003] In related technologies, existing simulation scene generation methods require manual construction of standard scenes based on pre-defined test requirements. These methods not only incur high manual costs and overall testing costs, but also suffer from low efficiency. Furthermore, they struggle to construct critical scenes to uncover key issues in autonomous driving systems. Summary of the Invention

[0004] To overcome the problems existing in related technologies, this disclosure provides a method, apparatus, device and storage medium for automatically constructing autonomous driving test scenarios.

[0005] According to a first aspect of the present disclosure, a method for automatically constructing autonomous driving test scenarios is provided, comprising:

[0006] Construct a test scenario, dynamically configure the configuration parameters of the test scenario to obtain multiple case scenarios, and add the case scenarios to the training queue;

[0007] The case scenarios in the training queue are run sequentially to obtain the running result of the main vehicle in the current case scenario. It is determined whether the running result has a preset behavior. If the preset behavior exists, the current case scenario is recorded in the first scenario library. If the preset behavior does not exist, the current case scenario is recorded in the second scenario library for the next step of judgment.

[0008] Obtain the running results of the case scenarios in the second scenario library, determine whether the running results have generated a preset feedback behavior, and trigger the case scenario corresponding to the judgment result to perform enhanced testing or delete it based on the judgment result;

[0009] According to a second aspect of the present disclosure, an automatic construction device for autonomous driving test scenarios is provided, comprising:

[0010] Scene configuration module: Used to build test scenarios, dynamically configure the configuration parameters of the test scenarios to obtain multiple case scenarios, and add the case scenarios to the training queue;

[0011] First judgment module: used to sequentially run the case scenarios in the training queue, obtain the running result of the main vehicle in the current case scenario, and determine whether the running result has a preset behavior. If the preset behavior exists, the current case scenario is recorded in the first scenario library. If the preset behavior does not exist, the current case scenario is recorded in the second scenario library for further judgment.

[0012] The second judgment module is used to obtain the running results of the case scenarios in the second scenario library, determine whether the running results have generated a preset feedback behavior, and trigger the case scenario corresponding to the judgment result to perform enhanced testing or deletion based on the judgment result.

[0013] According to a third aspect of the present disclosure, an electronic device is provided, comprising: a processor; a memory for storing executable instructions of the processor; the processor being configured to read the executable instructions from the memory and execute the instructions to implement the automatic construction method for autonomous driving test scenarios provided in the first aspect of the present disclosure.

[0014] According to a fourth aspect of the present disclosure, a computer-readable storage medium is provided, having stored thereon computer program instructions that, when executed by a processor, implement the steps of the automatic construction method for autonomous driving test scenarios provided in the first aspect of the present disclosure.

[0015] The technical solutions provided by the embodiments of this disclosure may include the following beneficial effects:

[0016] Based on pre-set standard scenarios, it can automatically build a large number of different case scenarios by automatically generating different scenario configuration items and parameters. The case scenarios cover more complex environments, simulate more scenario operation conditions, build key scenarios, discover more safety issues of autonomous driving systems, improve test results, and reduce test costs. By detecting vehicle running trajectory, it can discover case scenarios that may become key scenarios, making scenario testing more comprehensive, making it easier to discover autonomous driving safety issues, improving scenario generation efficiency and quality, improving test feasibility, reducing costs, and avoiding waste of resources and costs.

[0017] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0018] The accompanying drawings, which are incorporated in and form a part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure.

[0019] Figure 1 This is a flowchart illustrating an automatic construction method for autonomous driving test scenarios according to an exemplary embodiment.

[0020] Figure 2 This is an overall flowchart illustrating an automatic construction method for autonomous driving test scenarios according to an exemplary embodiment.

[0021] Figure 3 This is a block diagram illustrating an automatic construction device for autonomous driving test scenarios according to an exemplary embodiment.

[0022] Figure 4 This is a block diagram illustrating an electronic device according to an exemplary embodiment. Detailed Implementation

[0023] The exemplary embodiments will now be described in detail with reference to the accompanying drawings.

[0024] It should be noted that the relevant embodiments and accompanying drawings are only for describing and illustrating exemplary embodiments provided by this disclosure, and not all embodiments of this disclosure, nor should this disclosure be understood to be limited to the relevant exemplary embodiments.

[0025] It should be noted that the terms "first," "second," etc., used in this disclosure are only used to distinguish different steps, devices, or modules. These terms do not represent any specific technical meaning, nor do they indicate any order or interdependence between them.

[0026] It should be noted that the terms “a,” “a plurality of,” and “at least one” used in this disclosure are illustrative rather than restrictive. Unless otherwise expressly indicated in the context, they should be understood as “one or more.”

[0027] It should be noted that the term "and / or" used in this disclosure is used to describe the relationship between related objects, and generally indicates that there are at least three relationships. For example, A and / or B can at least indicate: the existence of A alone, the existence of both A and B, and the existence of B alone.

[0028] It should be noted that the various steps described in the method embodiments of this disclosure may be performed in different orders and / or in parallel. Unless otherwise specified, the scope of this disclosure is not limited to the order in which the steps are described in the relevant embodiments.

[0029] It should be noted that all actions involving the acquisition of signals, information, or data in this disclosure are carried out in compliance with the relevant data protection laws and policies of the country where the location is situated, and with authorization from the owner of the relevant device.

[0030] Exemplary methods

[0031] Figure 1 This is a flowchart illustrating an automatic construction method for autonomous driving test scenarios according to an exemplary embodiment, such as... Figure 1 As shown, the method for automatically constructing autonomous driving test scenarios includes the following steps.

[0032] In step S110, a test scenario is constructed. Based on the test scenario, the configuration parameters of the test scenario are dynamically configured to obtain multiple case scenarios, and the case scenarios are added to the training queue.

[0033] Conducting autonomous driving simulation testing requires building a large number of test scenarios, and it's essential to construct as many boundary and extreme scenarios as possible. To automate the construction of more valuable scenarios, a standard scenario is first built, or a pre-built standard scenario can be used as the first test scenario; among these,

[0034] The standard scenarios can be set up on straight sections, curves, and slopes of urban roads, highways, etc., as well as intersections, roundabouts, and highway entrance / exit ramps.

[0035] The first test scenario is dynamically configured based on preset configuration items, i.e., the preset configuration items are randomly combined and added to the first test scenario. The preset configuration items include at least: traffic participant behavior, traffic lights, weather conditions, road conditions, and vehicle headlight intensity. Traffic participant behavior includes at least: the position, speed, and direction of travel of other vehicles, and the position, speed, and direction of movement of pedestrians. Weather conditions include at least: sunny, rainy, and snowy.

[0036] The weather conditions include: day, temperature, ambient brightness, and visibility distance; road conditions include at least: puddles, icy surfaces, uneven surfaces, and road friction; vehicle headlight intensity includes at least: no lights on, high beams on, low beams on, turn signals on, and hazard lights on.

[0037] Within the preset parameter range of the configuration items, the parameters of the combined configuration items in the first test scenario are randomly assigned values ​​to generate multiple case scenarios based on the first test scenario. The generated cases...

[0038] The scenarios are sequentially placed into a training queue Q1, serving as training case 5 for autonomous driving simulation testing.

[0039] In step S120, the case scenarios in the training queue are run sequentially to obtain the running result of the main vehicle in the current case scenario. It is determined whether the running result has a preset behavior. If the preset behavior exists, the current case scenario is recorded in the first scenario library. If the preset behavior does not exist, the current case scenario is recorded in the second scenario library for the next step of judgment.

[0040] 0 sequentially retrieves case scenarios from training queue Q1, calls the autonomous driving system algorithm to be tested, and performs autonomous driving simulation testing in the case scenarios;

[0041] After the case scenario execution is completed, the execution results of the main vehicle in the current case scenario are obtained. The execution results include at least: the driving trajectory of the main vehicle, and whether the main vehicle interacted with other traffic participants during its driving process.

[0042] In the event of a collision, it is determined whether the main vehicle violated any traffic regulations during its journey; other traffic participants include at least other vehicles besides the main vehicle and pedestrians in the scenario; the violations of traffic regulations include speeding, failing to follow road signs, failing to follow road markings, failing to follow traffic signals, and failing to yield to pedestrians.

[0043] If the master vehicle's operation results show any unsafe behaviors such as collisions or traffic violations, the current case scenario is designated as a key scenario for the autonomous driving system algorithm and recorded in the first scenario library. If the master vehicle's operation results do not show any unsafe behaviors, the current case scenario is recorded in the second scenario library. It should be noted that the case scenarios recorded in the second scenario library include the configuration parameters and operation results of this scenario. The driving trajectory in the master vehicle's operation results of the case scenarios recorded in the second scenario library will be further evaluated.

[0044] In step S130, the running results of the case scenarios in the second scenario library are obtained, it is determined whether the running results have generated a preset feedback behavior, and the process of enhancing or deleting the case scenario corresponding to the judgment result is triggered based on the judgment result.

[0045] Obtain the running results of the case scenarios in the second scenario library. The main focus is to obtain the driving trajectory of the main vehicle in the running results and to determine whether the driving trajectory of the main vehicle in the case scenario has generated a preset feedback behavior. The feedback behavior is a new driving trajectory. If a feedback behavior has been generated, it means that in the current case scenario, the main vehicle has generated a new driving trajectory that was not generated during the previous case scenario test.

[0046] The specific method for determining whether the main vehicle has generated a new driving trajectory in the current scenario is as follows:

[0047] The first test scenario is divided into multiple blocks of standard size, such as 1m×1m or 2m×2m blocks. The side length of the blocks is set according to the actual situation of the scenario. Each block is numbered sequentially to record the driving trajectory of the main vehicle.

[0048] When the main vehicle is simulated in the current case scenario, the block area numbers passed by the main vehicle are recorded sequentially as the driving trajectory of the main vehicle in the current case scenario. It should be noted that the driving trajectory recorded during the running of each case scenario needs to be compared with the driving trajectories recorded by all previously run case scenarios through the block area numbers.

[0049] If the block number recorded in the current case scenario appears for the first time in all the run case scenarios, it means that the main vehicle has passed through a new block in the current case scenario and generated a new driving trajectory.

[0050] If the running result produces the preset feedback behavior, that is, the main vehicle produces a new driving trajectory that was not generated during the previous case scenario test, then the current case scenario is retained, the configuration parameters of the current case scenario are dynamically reconfigured, and the new case scenario is put into the training queue Q1 to wait for the scenario test to be carried out again; if no new driving trajectory is generated, the current case is discarded.

[0051] The tests show that the more different blocks the main vehicle passes through, the larger the activity area covered by the driving trajectory. This makes it easier to identify scenarios prone to unsafe behaviors that could lead to safety issues. Therefore, when dynamically reconfiguring case scenarios that produce preset feedback behaviors based on the running results, key configuration parameters should be modified in a targeted manner. The specific method is as follows:

[0052] By comparing the configuration items and parameter values ​​of the current case scenario with those of case scenarios that did not generate new feedback, the key configuration items in the current case scenario that may cause the master vehicle to generate a new driving trajectory are identified. The parameter values ​​of the identified key configuration items are modified to generate a new case scenario, which is then added to the training queue Q1. For example, if the current case scenario is compared with previously run case scenarios, and a traffic participant has been added to the current scenario (which could be another pedestrian or other vehicle), causing the master vehicle to generate a new driving trajectory, then the key configuration item may be the traffic participant. Therefore, when reconfiguring the current case scenario, another traffic participant can be added, which may again change the vehicle's driving trajectory. The corresponding judgment and processing are then performed according to the method in step S120.

[0053] This exemplary embodiment illustrates the overall process of an automatic construction method for autonomous driving test scenarios as follows: Figure 2As shown, a basic scenario is constructed and dynamically configured into multiple different cases, which are placed in queue Q1. The cases in Q1 are run sequentially, and the running result of the main vehicle in the case is judged. If the main vehicle in the case has a collision or violation safety issue while driving, the case is recorded as a key scenario. If no collision or violation occurs, the running of the main vehicle in the case is judged again to see if there is any new feedback, that is, whether a new driving trajectory has appeared. If there is a new driving trajectory, the current case is reconfigured in a direction that is conducive to discovering the problem, that is, the key configuration parameters that may lead to new feedback are modified. The reconfigured case is put back into queue Q1, and the above running process is repeated until the case is recorded or is discarded because no new feedback is generated.

[0054] The technical solutions provided by the embodiments of this disclosure may include the following beneficial effects:

[0055] Based on pre-set standard scenarios, it can automatically build a large number of different case scenarios by automatically generating different scenario configuration items and parameters. The case scenarios cover more complex environments, simulate more scenario operation conditions, build key scenarios, discover more safety issues of autonomous driving systems, improve test results, and reduce test costs. By detecting vehicle running trajectory, it can discover case scenarios that may become key scenarios, making scenario testing more comprehensive, making it easier to discover autonomous driving safety issues, improving scenario generation efficiency and quality, improving test feasibility, reducing costs, and avoiding waste of resources and costs.

[0056] Exemplary device

[0057] Figure 3 This is a block diagram of an automatic construction device for autonomous driving test scenarios, according to an exemplary embodiment. (Refer to...) Figure 3 The device 300 includes a scene configuration module 310, a first judgment module 320, and a second judgment module 330.

[0058] The scenario configuration module 310 is used to construct a test scenario, dynamically configure the configuration parameters of the test scenario to obtain multiple case scenarios, and add the case scenarios to the training queue.

[0059] When conducting autonomous driving simulation tests, it is necessary to construct a large number of test scenarios, and to construct as many boundary scenarios and extreme scenarios as possible. In order to automatically construct more valuable scenarios, a standard scenario is first constructed, or a pre-set standard scenario can be used as the first test scenario; wherein, the standard scenario can be set in straight sections, curves and slopes, intersections, roundabouts and highway entrance and exit ramps of urban roads, highways, etc.

[0060] The scene configuration module 310 is specifically used for:

[0061] The preset configuration items are randomly combined and added to the test scenario; wherein the configuration items include at least: the behavior of traffic participants, traffic lights, weather conditions, road conditions, and vehicle headlight intensity; wherein the behavior of traffic participants includes at least: the position, speed, and direction of travel of other vehicles, and the position, speed, and direction of movement of pedestrians; the weather conditions include at least: sunny, rainy, snowy, weather temperature, ambient brightness, and road visibility distance; the road conditions include at least: puddles, icy roads, uneven road surfaces, and road friction; the vehicle headlight intensity includes at least: no lights, high beams on, low beams on, turn signals on, and hazard lights on.

[0062] Within the preset parameter range of the configuration items, the parameters of the combined configuration items are randomly assigned to obtain multiple case scenarios. The generated case scenarios are then placed into a training queue Q1 as training case scenarios for autonomous driving simulation testing.

[0063] The first judgment module 320 is used to sequentially run the case scenarios in the training queue, obtain the running result of the main vehicle in the current case scenario, and determine whether the running result has a preset unsafe behavior. If the unsafe behavior exists, the current case scenario is recorded in the first scenario library. If the unsafe behavior does not exist, the current case scenario is recorded in the second scenario library for further judgment.

[0064] Specifically, after the case scenario execution is completed, the first judgment module 320 obtains the running result of the main vehicle in the current case scenario. The running result includes at least: the driving trajectory of the main vehicle, whether the main vehicle collided with other traffic participants during its driving, and whether the main vehicle violated traffic regulations during its driving. Among them, other traffic participants include at least other vehicles besides the main vehicle and pedestrians in the case scenario. The violations of traffic regulations include speeding, failure to follow road signs, failure to follow road markings, failure to follow traffic signals, and failure to yield to pedestrians.

[0065] If the master vehicle's operation results show any unsafe behaviors such as collisions or traffic violations, the current case scenario is designated as a key scenario for the autonomous driving system algorithm and recorded in the first scenario library. If the master vehicle's operation results do not show any unsafe behaviors, the current case scenario is recorded in the second scenario library. It should be noted that the case scenarios recorded in the second scenario library include the configuration parameters and operation results of this scenario. The driving trajectory in the master vehicle's operation results of the case scenarios recorded in the second scenario library will be further evaluated.

[0066] The second judgment module 330 is used to obtain the running results of the case scenarios in the second scenario library, determine whether the running results have generated a preset feedback behavior, and trigger the process of enhancing or deleting the case scenario corresponding to the judgment result based on the judgment result.

[0067] The second judgment module 330 is specifically used for:

[0068] If the running result produces the preset feedback behavior, the configuration parameters of the current case scenario are dynamically reconfigured, and the new case scenario is added to the training queue for scenario testing again; if the running result does not produce the preset feedback behavior, the current case scenario is discarded.

[0069] The feedback behavior includes at least the following: the main vehicle generates a new driving trajectory in the current case scenario;

[0070] The method for determining that the main vehicle has generated a new driving trajectory in the current scenario is as follows:

[0071] In the first test scenario, multiple block areas are set up and numbered.

[0072] The block numbers passed by the main vehicle are recorded sequentially as the driving trajectory of the main vehicle in the current case scenario;

[0073] If the block number recorded for the first time appears in the block area recorded in the current case scenario, it means that the main vehicle has generated a new driving trajectory in the current case scenario.

[0074] For case scenarios that generate new driving trajectories, the configuration parameters are dynamically reconfigured. The specific method is as follows:

[0075] By comparing the configuration items and parameter values ​​of the current case scenario with those of case scenarios that did not generate new feedback, the key configuration items in the current case scenario that may cause the master vehicle to generate a new driving trajectory are identified. The parameter values ​​of the identified key configuration items are then modified to generate a new case scenario, which is then added to the training queue Q1. For example, comparing the current case scenario with previously run case scenarios, the current scenario adds a traffic participant, which could be another pedestrian or other vehicle, causing the master vehicle to generate a new driving trajectory. The key configuration item is likely the traffic participant. Therefore, when reconfiguring the current case scenario, adding another traffic participant may again change the vehicle's driving trajectory.

[0076] This exemplary embodiment is an exemplary device embodiment corresponding to the above exemplary method embodiment. The specific operation of each module can be understood with reference to the description of the exemplary method embodiment, and will not be repeated here.

[0077] Exemplary electronic devices

[0078] Figure 4 This is a block diagram illustrating an electronic device 400 according to an exemplary embodiment. The electronic device 400 may be a vehicle controller, an in-vehicle terminal, an in-vehicle computer, or other types of electronic devices.

[0079] Reference Figure 4 The electronic device 400 may include at least one processor 410 and a memory 420. The processor 410 can execute instructions stored in the memory 420. The processor 410 is communicatively connected to the memory 420 via a data bus. In addition to the memory 420, the processor 410 can also be communicatively connected to an input device 430, an output device 440, and a communication device 450 via the data bus.

[0080] Processor 410 can be any conventional processor, such as a commercially available CPU. The processor may also include, for example, a Graphics Processing Unit (GPU), a Field Programmable Gate Array (FPGA), a System on Chip (SOC), an Application Specific Integrated Circuit (ASIC), or a combination thereof.

[0081] The memory 420 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk or optical disk.

[0082] In this embodiment of the present disclosure, the memory 420 stores executable instructions, and the processor 410 can read the executable instructions from the memory 420 and execute the instructions to implement all or part of the steps of the automatic construction method for autonomous driving test scenarios described in any of the exemplary embodiments above.

[0083] Exemplary computer-readable storage media

[0084] In addition to the methods and apparatus described above, exemplary embodiments of this disclosure may also be a computer program product or a computer-readable storage medium storing the computer program product. The computer product includes computer program instructions that can be executed by a processor to implement all or part of the steps for automatically constructing the autonomous driving test scenario described in any of the methods of the exemplary embodiments above.

[0085] The computer program product can be written in any combination of one or more programming languages ​​to perform the operations of the embodiments of this application. The programming languages ​​include object-oriented programming languages ​​such as Java and C++, as well as conventional procedural programming languages ​​such as C or similar languages, and scripting languages ​​(e.g., Python). The program code can be executed entirely on the user's computing device, partially on the user's device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0086] The computer-readable storage medium may be any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of readable storage media include: static random access memory (SRAM) having one or more electrically connected wires, electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk or optical disk, or any suitable combination thereof.

[0087] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of this disclosure. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the following claims.

[0088] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.

Claims

1. A method for automatically constructing autonomous driving test scenarios, characterized in that, include: Construct a test scenario, dynamically configure the configuration parameters of the test scenario to obtain multiple case scenarios, and add the case scenarios to the training queue; The case scenarios in the training queue are run sequentially to obtain the running result of the main vehicle in the current case scenario. It is determined whether the running result has a preset behavior. If the preset behavior exists, the current case scenario is recorded in the first scenario library. If the preset behavior does not exist, the current case scenario is recorded in the second scenario library for the next step of judgment. Obtain the running results of the case scenarios in the second scenario library, determine whether the running results have generated a preset feedback behavior, and trigger the case scenario corresponding to the judgment result to perform enhanced testing or delete it based on the judgment result; The specific process for handling case scenarios based on the judgment result includes: If the running result produces the preset feedback behavior, the configuration parameters of the current case scenario are dynamically reconfigured, and the new case scenario is added to the training queue for scenario testing again; if the running result does not produce the preset feedback behavior, the current case scenario is discarded. The specific method for dynamically reconfiguring the configuration parameters of the current case scenario is as follows: Compare the configuration items and parameter values ​​of the current case with those of cases that have not generated new feedback to identify the key configuration items in the current case that are causing new feedback. Modify the parameter values ​​of the key configuration items to generate new cases; The feedback behavior includes at least the following: the main vehicle generates a new driving trajectory in the current case scenario; The method for determining that the main vehicle has generated a new driving trajectory in the current scenario is as follows: Multiple blocks are set up in the test scenario, and the blocks are numbered. The block numbers passed by the main vehicle are recorded sequentially as the driving trajectory of the main vehicle in the current case scenario; If the block number recorded for the first time appears in the block area recorded in the current case scenario, it means that the main vehicle has generated a new running trajectory in the current case scenario.

2. The method according to claim 1, characterized in that, The process of dynamically configuring the configuration parameters of the test scenario to obtain multiple case scenarios specifically includes: The preset configuration items are randomly combined and added to a standard test scenario; wherein the configuration items include at least: the behavior of traffic participants, traffic lights, weather conditions, road conditions, and vehicle light intensity; wherein the behavior of traffic participants includes at least: the position, speed, and direction of travel of other vehicles, and the position, speed, and direction of movement of pedestrians; Within the preset parameter range of the configuration items, the parameters of the combined configuration items are randomly assigned to obtain multiple case scenarios.

3. An automatic construction device for autonomous driving test scenarios, characterized in that, include: Scene configuration module: Used to build test scenarios, dynamically configure the configuration parameters of the test scenarios to obtain multiple case scenarios, and add the case scenarios to the training queue; First judgment module: used to sequentially run the case scenarios in the training queue, obtain the running result of the main vehicle in the current case scenario, and determine whether the running result has a preset behavior. If the preset behavior exists, the current case scenario is recorded in the first scenario library. If the preset behavior does not exist, the current case scenario is recorded in the second scenario library for further judgment. The second judgment module is used to obtain the running results of the case scenarios in the second scenario library, determine whether the running results have generated a preset feedback behavior, and trigger the case scenario corresponding to the judgment result to perform enhanced testing or deletion based on the judgment result. The second judgment module is specifically used for: If the running result produces the preset feedback behavior, the configuration parameters of the current case scenario are dynamically reconfigured, and the new case scenario is added to the training queue for scenario testing again; if the running result does not produce the preset feedback behavior, the current case scenario is discarded. The specific method for dynamically reconfiguring the configuration parameters of the current case scenario is as follows: Compare the configuration items and parameter values ​​of the current case with those of cases that have not generated new feedback to identify the key configuration items in the current case that are causing new feedback. Modify the parameter values ​​of the key configuration items to generate new cases; The feedback behavior includes at least the following: the main vehicle generates a new driving trajectory in the current case scenario; The second judgment module is further used for: Multiple block areas are set up in the test scenario, and the block areas are numbered. The block numbers passed by the main vehicle are recorded sequentially as the driving trajectory of the main vehicle in the current case scenario; If the block number recorded for the first time appears in the block area recorded in the current case scenario, it means that the main vehicle has generated a new running trajectory in the current case scenario.

4. The apparatus according to claim 3, characterized in that, The scenario configuration module is specifically used for: The preset configuration items are randomly combined and added to a standard test scenario; wherein the configuration items include at least: the behavior of traffic participants, traffic lights, weather conditions, road conditions, and vehicle light intensity; wherein the behavior of traffic participants includes at least: the position, speed, and direction of travel of other vehicles, and the position, speed, and direction of movement of pedestrians; Within the preset parameter range of the configuration items, the parameters of the combined configuration items are randomly assigned to obtain multiple case scenarios.

5. An electronic device, characterized in that, include: processor; Memory used to store the processor's executable instructions; The processor is configured to read the executable instructions from the memory and execute the instructions to implement the automatic construction method for autonomous driving test scenarios as described in any one of claims 1-2.

6. A computer-readable storage medium having computer program instructions stored thereon, characterized in that, When the program instructions are executed by the processor, they implement the steps of the automatic construction method for autonomous driving test scenarios as described in any one of claims 1-2.

Images