Method and system for testing automatic emergency braking system of vehicle

Abstract

The invention provides a vehicle automatic emergency braking system test method and system, and relates to the technical field of vehicle performance test.The method inputs actual driving data of a to-be-tested vehicle into a dummy model which is trained in advance based on machine learning, can accurately simulate the response, the traversing action and the speed of pedestrians facing the vehicle, and improves the performance of the to-be-tested vehicle. Therefore, the accuracy and efficiency of vehicle safety performance testing are remarkably improved, and powerful support is provided for vehicle design optimization and intelligent driving technology development.

Description

Technical Field

[0001] The present application relates to the technical field of vehicle performance testing, and in particular to a vehicle automatic emergency braking system testing method and system. Background Art

[0002] With the rapid development of intelligent driving technology, people's demand for driving safety is increasing. Automatic emergency braking system (AEB) is one of the key technologies to improve driving safety. The AEB system can automatically brake when a potential collision risk is detected to avoid or reduce the occurrence of traffic accidents. However, the test scenarios for pedestrians crossing in the current AEB test standards are often too simplified and cannot fully cover the complex and changeable traffic accident situations in real life.

[0003] In real life, pedestrians' behaviors when crossing the road are highly random and complex, including different reaction times, crossing speeds, walking trajectories, etc., which pose great challenges to the accurate judgment of the AEB system. Traditional testing methods are often based on preset and fixed test scenarios, which are difficult to truly reflect the actual behavior of pedestrians when facing vehicles, resulting in a large deviation between the test results and actual usage. Summary of the invention

[0004] In view of the above problems, the present application provides a vehicle automatic emergency braking system testing method, including the following contents:

[0005] In a first aspect, the present application provides a vehicle automatic emergency braking system testing method, the method comprising:

[0006] Obtaining actual driving data of the vehicle to be tested;

[0007] The actual driving data is input into a dummy model pre-trained based on machine learning; the dummy model is used to simulate the pedestrian's reaction, crossing action and speed when facing a vehicle;

[0008] Based on the actual driving data and the historical collision data in the dummy model pre-trained based on machine learning, the dummy model is controlled to perform avoidance or crossing actions according to the operation conditions of the vehicle.

[0009] Optionally, before the dummy model is trained, the method further includes:

[0010] Obtaining pedestrian crossing information in a traffic accident where a vehicle cannot brake urgently and thus collide with a pedestrian crossing the vehicle, wherein the pedestrian crossing information includes crossing speed, waving speed, and collision position;

[0011] Acquire vehicle information corresponding to the pedestrian crossing information, including vehicle speed and vehicle model;

[0012] Acquire environmental information corresponding to the pedestrian crossing information, including the lateral distance and longitudinal distance between the vehicle and the pedestrian, and surrounding environment occlusion information;

[0013] The pedestrian crossing information, vehicle information and environmental information are stored in a resource library and feature marked;

[0014] A pedestrian crossing collision model is established based on machine learning. The pedestrian's reaction, crossing action and speed when facing a vehicle are simulated to train the dummy model.

[0015] Optionally, the obtaining actual driving data of the vehicle to be tested includes:

[0016] The real-time speed, position and heading angle information of the vehicle to be tested are obtained in real time using on-board sensors or external monitoring systems.

[0017] Optionally, the method further includes:

[0018] After the dummy model triggers the crossing action, the vehicle information, collision information of the vehicle to be tested and the crossing information of the dummy model are recorded as the test result.

[0019] Optionally, the method further includes:

[0020] Selecting functions for the dummy model, including a false triggering function, which is used to control the dummy model to simulate a test scenario in which a vehicle is falsely triggered due to a pedestrian's sudden stop, turn, or other actions when crossing the vehicle;

[0021] Direct trigger function, used to control the dummy model to directly perform a traversing action to trigger the vehicle's automatic emergency braking AEB system;

[0022] The parameter matching function with the preset relevant standard and regulatory database is used to control the dummy model to automatically match the corresponding regulatory parameters based on the pre-entered specific regulatory certification requirements, and make corresponding crossing choices based on vehicle speed and other conditions;

[0023] Generate different test scenarios based on the selected functions and parameter matching results with the preset standard and regulatory database.

[0024] In a second aspect, the present application provides a vehicle automatic emergency braking system test system, the system comprising:

[0025] An acquisition unit, used to acquire actual driving data of the vehicle to be tested;

[0026] A data processing unit, used to input the actual driving data into a dummy model pre-trained based on machine learning; the dummy model is used to simulate the pedestrian's reaction, crossing action and speed when facing a vehicle;

[0027] A control unit is used to control the dummy model to perform avoidance or crossing actions according to the vehicle's operating conditions based on the actual driving data and the historical collision data in the dummy model that has been pre-trained based on machine learning.

[0028] Optionally, the acquisition unit is further used to, before the dummy model is trained, acquire pedestrian crossing information in a traffic accident where a vehicle cannot brake urgently and thus causes a collision, wherein the pedestrian crossing information includes a crossing speed, a waving speed, and a collision position;

[0029] Acquire vehicle information corresponding to the pedestrian crossing information, including vehicle speed and vehicle model;

[0030] Acquire environmental information corresponding to the pedestrian crossing information, including the lateral distance and longitudinal distance between the vehicle and the pedestrian, and surrounding environment occlusion information;

[0031] The pedestrian crossing information, vehicle information and environmental information are stored in a resource library and feature marked;

[0032] A pedestrian crossing collision model is established based on machine learning. The pedestrian's reaction, crossing action and speed when facing a vehicle are simulated to train the dummy model.

[0033] Optionally, the acquisition unit acquires the actual driving data of the vehicle to be tested by using a vehicle-mounted sensor or an external monitoring system to acquire the real-time vehicle speed, position and heading angle information of the vehicle to be tested in real time.

[0034] Optionally, the data processing unit is further used to record the vehicle information, collision information of the vehicle to be tested and the dummy model traversing information as the test result after the dummy model triggers the traversing action.

[0035] Optionally, the system further includes a function selection unit for selecting functions of the dummy model, wherein the functions include a false triggering function for controlling the dummy model to simulate a test scenario in which a vehicle is falsely triggered due to a pedestrian's sudden stop, turn, or other actions when crossing the vehicle;

[0036] Direct trigger function, used to control the dummy model to directly perform a traversing action to trigger the vehicle's automatic emergency braking AEB system;

[0037] The parameter matching function with the preset relevant standard and regulatory database is used to control the dummy model to automatically match the corresponding regulatory parameters based on the pre-entered specific regulatory certification requirements, and make corresponding crossing choices based on vehicle speed and other conditions;

[0038] Generate different test scenarios based on the selected functions and parameter matching results with the preset standard and regulatory database.

[0039] In a third aspect, the present application provides a device comprising a memory and a processor, wherein the memory is used to store instructions or codes, and the processor is used to execute the instructions or codes so that the device executes the vehicle automatic emergency braking system testing method introduced in any implementation of the first aspect.

[0040] In a fourth aspect, the present application provides a computer-readable storage medium, in which a code is stored. When the code is executed, the device executing the code implements the vehicle automatic emergency braking system testing method introduced in any implementation method of the aforementioned first aspect.

[0041] The present application provides a vehicle automatic emergency braking system testing method. When executing the method, the actual driving data of the vehicle to be tested is first obtained, and then the actual driving data is input into a dummy model that has been pre-trained based on machine learning. The dummy model is used to simulate the reaction, crossing action and speed of pedestrians when facing the vehicle. Based on the actual driving data and the historical collision data in the dummy model that has been pre-trained based on machine learning, the dummy model is controlled to perform avoidance or crossing actions according to the operation of the vehicle. By inputting the actual driving data of the vehicle to be tested into a dummy model that has been pre-trained based on machine learning through this method, the reaction, crossing action and speed of pedestrians when facing the vehicle can be accurately simulated, thereby significantly improving the accuracy and efficiency of vehicle safety performance testing, and providing strong support for vehicle design optimization and the development of intelligent driving technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions in this embodiment or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0043] Figure 1 A flow chart of a vehicle automatic emergency braking system testing method provided in an embodiment of the present application;

[0044] Figure 2 A schematic diagram of the structure of a vehicle automatic emergency braking system test system provided in an embodiment of the present application. DETAILED DESCRIPTION

[0045] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0046] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant laws, regulations and standards of relevant countries and regions.

[0047] Figure 1 A flow chart of a vehicle automatic emergency braking system testing method provided in an embodiment of the present application. Figure 1 As shown, the vehicle automatic emergency braking system testing method provided by the embodiment of the present application may include:

[0048] S101. Acquire actual driving data of a vehicle to be tested.

[0049] Vehicles are usually equipped with a variety of sensors, such as speed sensors, GPS sensors (global positioning system sensors), etc. These sensors can monitor and record the speed, position and heading angle of the vehicle (i.e. the direction of the vehicle's travel) in real time. The speed sensor can measure the rotation speed of the wheels or the acceleration of the vehicle to calculate the actual speed of the vehicle. The GPS sensor can receive signals from multiple satellites and determine the three-dimensional position (longitude, latitude and altitude) of the vehicle by calculating the propagation time of these signals. The heading angle of the vehicle can usually be measured by analyzing the motion trajectory in the GPS data or using a dedicated heading sensor. The heading angle represents the angle between the direction of the vehicle's travel and a reference direction (such as due north).

[0050] In addition to on-board sensors, external monitoring systems (such as traffic monitoring cameras, radar systems, etc.) can also be used to monitor and record vehicle driving data. This method is usually used in scenarios such as traffic management and security monitoring, and can also be used for vehicle testing.

[0051] Collecting this real-time data is critical to evaluating the performance, safety, and effectiveness of the vehicle's driver assistance systems. During the vehicle testing phase, this data can help engineers identify potential problems, optimize the vehicle's design and control systems, and ensure that the vehicle performs well under a variety of driving conditions.

[0052] S102, inputting the actual driving data into a dummy model that has been pre-trained based on machine learning; the dummy model is used to simulate the pedestrian's reaction, crossing action and speed when facing a vehicle.

[0053] The actual driving data includes real-time data such as speed, position and heading angle obtained from the vehicle, which reflects the driving status of the vehicle at a specific moment.

[0054] The pre-trained dummy model is trained using machine learning techniques to simulate pedestrians’ various reactions, crossing movements, and walking speeds when facing vehicles. The training of the model usually involves the collection and analysis of a large amount of pedestrian behavior data, and the use of this data to optimize the model’s predictive capabilities.

[0055] The actual driving data is fed into the dummy model, which then simulates the likely behavior of pedestrians based on the data. For example, if the vehicle is approaching a location where a pedestrian might cross, the model can predict whether the pedestrian will choose to cross, when, and at what speed.

[0056] By using a dummy model to simulate real-world pedestrian behavior, the performance of the vehicle's automatic emergency braking (AEB) system or other driver assistance systems can be evaluated in complex scenarios such as pedestrians crossing the road. By combining actual driving data with simulated pedestrian behavior, a test environment close to the real world can be created, which is crucial for verifying the safety and reliability of vehicle systems.

[0057] S103: Based on the actual driving data and the historical collision data in the dummy model pre-trained based on machine learning, control the dummy model to perform an avoidance or crossing action according to the operation status of the vehicle.

[0058] Combine the actual driving data of the vehicle to be tested (such as speed, position and heading angle) with the historical collision data in the dummy model. The historical collision data may contain pedestrian behavior patterns in a variety of scenarios, such as pedestrian reactions and actions at different speeds and distances. Based on this data, the system or model will make a decision to determine whether the dummy model should perform an evasive action or a crossing action. This decision-making process may involve complex algorithms and logic to ensure that the simulated pedestrian behavior is both realistic and representative. Once the decision is made, the dummy model will perform the corresponding action based on the decision result, such as stopping to avoid the vehicle or continuing to cross the road.

[0059] When the dummy model triggers the crossing action, it usually means that a potential collision scenario has occurred or is about to occur. At this time, the system will record the vehicle information (such as model, speed, etc.), collision information (such as collision location, collision intensity, etc.) of the vehicle to be tested, and the crossing information of the dummy model (such as crossing speed, crossing location, etc.). This information together constitutes a complete record of the test results for subsequent analysis and evaluation.

[0060] By simulating real-world pedestrian behavior, the performance of the vehicle under test in potential collision scenarios is evaluated. By recording and analyzing the test results, the vehicle's response and safety in complex scenarios such as pedestrians crossing can be understood, thus providing valuable feedback for vehicle design and improvement.

[0061] The above embodiment introduces the process of testing the automatic emergency braking system of a vehicle based on a pre-trained dummy model and the actual driving data of the vehicle to be tested. In the implementation process of the above method, before the dummy model is trained, the method also includes: obtaining pedestrian crossing information in a traffic accident where the vehicle cannot perform emergency braking due to the pedestrian crossing information causing a collision, the pedestrian crossing information including crossing speed, waving speed, and collision position; obtaining vehicle information corresponding to the pedestrian crossing information, including vehicle speed and vehicle model; obtaining environmental information corresponding to the pedestrian crossing information, including the lateral distance and longitudinal distance between the vehicle and the pedestrian, and surrounding environment occlusion information; storing the pedestrian crossing information, vehicle information, and environmental information in a resource library and performing feature marking; establishing a pedestrian crossing collision model based on machine learning, and training the dummy model by simulating and learning the pedestrian's reaction, crossing action, and speed when facing a vehicle.

[0062] To obtain pedestrian crossing information in traffic accidents, we collect collision cases in which pedestrians cross vehicles in real traffic accidents and the vehicles fail to brake urgently. Pedestrian crossing information mainly includes the crossing speed of pedestrians, i.e. the walking speed of pedestrians when crossing the road, the waving speed (pedestrians may wave their hands to indicate that they want to cross), and the collision position (the specific position where the pedestrian and the vehicle come into contact).

[0063] The vehicle information corresponding to the pedestrian crossing information is obtained to understand the specific state of the vehicle when the collision occurs. The vehicle information includes the vehicle speed (the speed at the moment of collision) and the vehicle model (different models may have different braking performance and field of view).

[0064] The environmental information corresponding to the pedestrian crossing information is obtained to analyze the environmental factors when the collision occurs. The environmental information includes the lateral distance between the vehicle and the pedestrian (i.e., the vertical distance between the vehicle and the pedestrian's crossing path), the longitudinal distance (i.e., the relative distance between the vehicle and the pedestrian along the road direction), and the surrounding environmental occlusion information (such as trees, buildings, etc. that may block the driver's line of sight).

[0065] Storing the pedestrian crossing information, vehicle information and environmental information in a resource library can organize the collected information into a usable data set. Specifically, the pedestrian crossing information, vehicle information and environmental information are stored in a resource library, and each piece of data is marked with features for subsequent analysis and model training.

[0066] After completing the above preparations, a pedestrian crossing collision model is established based on machine learning, and a model that can simulate pedestrian crossing behavior is established using machine learning technology. Through simulation, learn the pedestrian's reaction when facing a vehicle (such as whether to choose to cross, when to start crossing), crossing action (such as walking speed, walking trajectory) and speed, and how these factors affect the occurrence of collisions. This model will be trained based on the data in the resource library to optimize its prediction ability. Further train the dummy model so that the dummy model can simulate pedestrian behavior in the real world. Specifically, the behavior of the dummy model is guided by using the trained pedestrian crossing collision model. By continuously adjusting and optimizing the parameters and logic of the dummy model, it can more accurately simulate the reactions and actions of pedestrians when facing vehicles.

[0067] In summary, these steps together constitute a complete dummy model training process. By collecting and analyzing data from real traffic accidents and building a model using machine learning technology, a dummy model that can simulate real-world pedestrian behavior can be trained. This model will be used in subsequent vehicle tests to evaluate the performance and safety of the vehicle when facing complex scenarios such as pedestrians crossing the road.

[0068] In one implementation of the embodiment of the present application, the dummy model can support different function selections to achieve more flexible testing of the AEB system. The optional functions of the dummy model include false triggering function, direct triggering function and parameter matching function with the preset relevant standard and regulatory database. In order to simulate the scenario in the real world where the vehicle may be falsely triggered due to pedestrians crossing, suddenly stopping, turning, etc., a false triggering function is designed for the dummy model to control the dummy model to simulate the test scenario where the vehicle is falsely triggered due to pedestrians crossing, suddenly stopping, turning, etc. By controlling the dummy model to perform these specific actions, the vehicle's sensors and control system can be triggered, thereby evaluating the vehicle's response and performance when facing such complex and unexpected pedestrian behavior.

[0069] In order to directly test the vehicle's automatic emergency braking (AEB) system, a direct trigger function is designed for the dummy model, which is used to control the dummy model to directly perform a traversing action to trigger the vehicle's automatic emergency braking AEB system. By controlling the dummy model to directly perform a traversing action, the vehicle's AEB system can be triggered and its response speed, braking effect, and overall safety can be evaluated.

[0070] In order to ensure that the test process meets specific regulatory certification requirements, a function for matching parameters with a preset standard regulatory database is designed for the dummy model. The function of matching parameters with a preset relevant standard regulatory database is used to control the dummy model to automatically match the corresponding regulatory parameters based on the pre-entered specific regulatory certification requirements, and make corresponding crossing choices based on vehicle speed and other conditions; different test scenarios are generated based on the selected function and the parameter matching results with the preset standard regulatory database. Based on the pre-entered specific regulatory certification requirements, the dummy model can automatically match the corresponding regulatory parameters. For example, according to the regulatory requirements for the vehicle's reaction time or braking distance to pedestrians crossing at different speeds, the dummy model can make corresponding crossing choices.

[0071] After completing the function selection of the dummy model, you can use these functions to generate different test scenarios according to the test requirements and conduct actual vehicle tests. By collecting and analyzing the test results, you can accurately evaluate the performance of the vehicle and provide valuable feedback for vehicle design and improvement.

[0072] The above are some specific implementations of a vehicle automatic emergency braking system test method provided in the embodiment of the present application. Based on this, the present application also provides a corresponding system. The system provided in the embodiment of the present application will be introduced from the perspective of functional modularization.

[0073] Figure 2 A schematic diagram of the structure of a vehicle automatic emergency braking system test system provided in an embodiment of the present application. Figure 2 As shown, the vehicle automatic emergency braking system test system 200 provided in the embodiment of the present application includes:

[0074] An acquisition unit 210 is used to acquire actual driving data of the vehicle to be tested;

[0075] The data processing unit 220 is used to input the actual driving data into a dummy model that has been pre-trained based on machine learning; the dummy model is used to simulate the pedestrian's reaction, crossing action and speed when facing a vehicle;

[0076] The control unit 230 is used to control the dummy model to perform avoidance or crossing actions according to the vehicle's operating conditions based on the actual driving data and the historical collision data in the dummy model pre-trained based on machine learning.

[0077] In an implementation of the embodiment of the present application, the acquisition unit is further used to, before the dummy model is trained, acquire pedestrian crossing information in a traffic accident where a vehicle cannot perform emergency braking and thus collide with the pedestrian crossing vehicle, wherein the pedestrian crossing information includes a crossing speed, a hand waving speed, and a collision position;

[0078] Acquire vehicle information corresponding to the pedestrian crossing information, including vehicle speed and vehicle model;

[0079] Acquire environmental information corresponding to the pedestrian crossing information, including the lateral distance and longitudinal distance between the vehicle and the pedestrian, and surrounding environment occlusion information;

[0080] The pedestrian crossing information, vehicle information and environmental information are stored in a resource library and feature marked;

[0081] A pedestrian crossing collision model is established based on machine learning. The pedestrian's reaction, crossing action and speed when facing a vehicle are simulated to train the dummy model.

[0082] In one implementation of the embodiment of the present application, the acquisition unit acquires the actual driving data of the vehicle to be tested, including using an on-board sensor or an external monitoring system to acquire the real-time vehicle speed, position and heading angle information of the vehicle to be tested in real time.

[0083] In an implementation of the embodiment of the present application, the data processing unit is further used to record the vehicle information, collision information and crossing information of the vehicle to be tested as the test result after the dummy model triggers the crossing action.

[0084] In one implementation of the embodiment of the present application, the system further includes a function selection unit for selecting functions for the dummy model, wherein the functions include a false triggering function for controlling the dummy model to simulate a test scenario in which a vehicle is falsely triggered due to a pedestrian's sudden stop, turn, or other actions when crossing;

[0085] Direct trigger function, used to control the dummy model to directly perform a traversing action to trigger the vehicle's automatic emergency braking AEB system;

[0086] The parameter matching function with the preset relevant standard and regulatory database is used to control the dummy model to automatically match the corresponding regulatory parameters based on the pre-entered specific regulatory certification requirements, and make corresponding crossing choices based on vehicle speed and other conditions;

[0087] Generate different test scenarios based on the selected functions and parameter matching results with the preset standard and regulatory database.

[0088] The embodiments of the present application also provide corresponding devices and computer storage media for implementing the solutions provided by the embodiments of the present application.

[0089] The device includes a memory and a processor, the memory is used to store instructions or codes, and the processor is used to execute the instructions or codes so that the device executes the method described in any embodiment of the present application.

[0090] The computer storage medium stores codes, and when the codes are executed, a device executing the codes implements the method described in any embodiment of the present application.

[0091] Through the description of the above implementation methods, it can be known that those skilled in the art can clearly understand that all or part of the steps in the above-mentioned embodiment method can be implemented by means of software plus a general hardware platform. Based on such an understanding, the technical solution of the present application can be embodied in the form of a software product, which can be stored in a storage medium, such as a read-only memory (ROM) / RAM, a magnetic disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network communication device such as a router) to execute the methods described in each embodiment of the present application or some parts of the embodiments.

[0092] It is understandable that in the specific implementation of the present application, the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved, when the above embodiments of the present application are applied to specific products or technologies, need to obtain user permission or consent, and the collection, use and processing of relevant data need to comply with relevant laws, regulations and standards of relevant countries and regions.

[0093] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0094] It should also be noted that the various embodiments in this specification are described in a progressive manner, and the same and similar parts between the various embodiments can refer to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device and system embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can refer to the partial description of the method embodiments. The device and system embodiments described above are merely schematic, in which the units described as separate components may or may not be physically separated, and the components indicated as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0095] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A vehicle automatic emergency braking system testing method, characterized in that: The method comprises: Obtaining actual driving data of the vehicle to be tested; The actual driving data is input into a dummy model pre-trained based on machine learning; the dummy model is used to simulate the pedestrian's reaction, crossing action and speed when facing a vehicle; Based on the actual driving data and the historical collision data in the dummy model pre-trained based on machine learning, the dummy model is controlled to perform avoidance or crossing actions according to the operation conditions of the vehicle.

2. The method according to claim 1, characterized in that Before the dummy model is trained, the method further includes: Obtaining pedestrian crossing information in a traffic accident where a vehicle cannot brake urgently and thus collide with a pedestrian crossing the vehicle, wherein the pedestrian crossing information includes crossing speed, waving speed, and collision position; Acquire vehicle information corresponding to the pedestrian crossing information, including vehicle speed and vehicle model; Acquire environmental information corresponding to the pedestrian crossing information, including the lateral distance and longitudinal distance between the vehicle and the pedestrian, and surrounding environment occlusion information; The pedestrian crossing information, vehicle information and environmental information are stored in a resource library and feature marked; A pedestrian crossing collision model is established based on machine learning. The pedestrian's reaction, crossing action and speed when facing a vehicle are simulated to train the dummy model.

3. The method according to claim 1, characterized in that The obtaining of actual driving data of the vehicle to be tested comprises: The real-time speed, position and heading angle information of the vehicle to be tested are obtained in real time using on-board sensors or external monitoring systems.

4. The method according to claim 1, characterized in that: The method further comprises: After the dummy model triggers the crossing action, the vehicle information, collision information of the vehicle to be tested and the crossing information of the dummy model are recorded as the test result.

5. The method according to claim 1, characterized in that The method further comprises: Selecting functions for the dummy model, including a false triggering function, which is used to control the dummy model to simulate a test scenario in which a vehicle is falsely triggered due to a pedestrian's sudden stop, turn, or other actions when crossing the vehicle; Direct trigger function, used to control the dummy model to directly perform a traversing action to trigger the vehicle's automatic emergency braking AEB system; The parameter matching function with the preset relevant standard and regulatory database is used to control the dummy model to automatically match the corresponding regulatory parameters based on the pre-entered specific regulatory certification requirements, and make corresponding crossing choices based on vehicle speed and other conditions; Generate different test scenarios based on the selected functions and parameter matching results with the preset standard and regulatory database.

6. A vehicle automatic emergency braking system test system, characterized in that: The system comprises: An acquisition unit, used to acquire actual driving data of the vehicle to be tested; A data processing unit, used to input the actual driving data into a dummy model pre-trained based on machine learning; the dummy model is used to simulate the pedestrian's reaction, crossing action and speed when facing a vehicle; A control unit is used to control the dummy model to perform avoidance or crossing actions according to the vehicle's operating conditions based on the actual driving data and the historical collision data in the dummy model that has been pre-trained based on machine learning.

7. The system according to claim 6, characterized in that The data processing unit is further configured to record vehicle information, collision information of the vehicle to be tested and crossing information of the dummy model as test results after the dummy model triggers a crossing action.

8. The system according to claim 6, characterized in that The system further includes a function selection unit for selecting functions of the dummy model, wherein the functions include a false triggering function for controlling the dummy model to simulate a test scenario in which a vehicle is falsely triggered due to a pedestrian's sudden stop, turn, or other actions when crossing the vehicle; Direct trigger function, used to control the dummy model to directly perform a traversing action to trigger the vehicle's automatic emergency braking AEB system; The parameter matching function with the preset relevant standard and regulatory database is used to control the dummy model to automatically match the corresponding regulatory parameters based on the pre-entered specific regulatory certification requirements, and make corresponding crossing choices based on vehicle speed and other conditions; Generate different test scenarios based on the selected functions and parameter matching results with the preset standard and regulatory database.

9. A computing device, characterized in that The computing device includes: a memory and a processor; The memory is used to store computer programs; The processor is configured to implement the method according to any one of claims 1 to 5 when executing the computer program.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.

Images