Test method and system of vehicle AEB system

Abstract

The invention provides a test method and system of a vehicle AEB system, and belongs to the technical field of vehicle test. The method comprises the steps of generating a virtual test scene; collecting simulation video data in the virtual test scene; setting the scene recognition precision of the virtual vehicle, and simulating the message information of the virtual vehicle in the virtual test scene; simulation data including the simulation video data and the message information are sent to an AEB system of the tested vehicle, and a control signal, namely the braking deceleration, of the virtual vehicle is output; converting the brake deceleration into a brake pedal signal; inputting the brake pedal signal into a dynamical model of the test vehicle, and outputting motion information of the test vehicle; updating message information of the virtual vehicle in the virtual test scene according to the motion information; and comparing the updated message information with the standard message information in the braking state, and outputting a test result of the AEB system. The AEB system test accuracy is improved, the test period is shortened, and the test cost is reduced.

Description

Technical Field

[0001] The present invention relates to the field of vehicle testing technology, and in particular to a testing method and system for a vehicle AEB system. Background Art

[0002] The AEB system is an active safety technology that can reduce the vehicle's speed and avoid collision accidents when it detects a risk of collision with the vehicle in front. Vehicles often encounter complex traffic environments during driving, such as road congestion and pedestrians crossing. The AEB system can monitor the road conditions ahead in real time, issue warnings in time and take measures to avoid collision accidents caused by the driver's untimely response. Vehicles usually need to travel under various road conditions, including highways, urban roads, rural roads, etc. The AEB system can automatically adjust the braking strategy and braking force according to different road conditions and vehicle speeds to ensure effective safety protection in various situations. The current vehicle is equipped with an AEB system, which has become an inevitable trend in the industry.

[0003] The vehicle AEB system is extremely important, so rapid testing and verification of the vehicle AEB system has become a difficult problem in the industry. The mainstream testing method in the industry is generally real-vehicle testing, which is mainly based on closed-site testing and open-road testing. Real-vehicle testing has many defects: (1) The testing efficiency is low. It is necessary to find a site, install and debug equipment, which generally takes at least three days, and it is difficult to meet the development and iteration speed of the vehicle AEB algorithm; (2) The test process of the real-vehicle test is difficult to fully reproduce, that is, the test process cannot be traced, and the lighting, temperature, wind direction, etc. at the time of the test are difficult to reproduce; (3) When facing dangerous scenes, the real-vehicle test has high testing risks, especially high-speed scenes, and more extreme cuts, etc., which cannot be achieved by real-vehicle testing; (4) The cost of real-vehicle testing is high, often more than one million, and there is a risk of damage; Therefore, the industry is in urgent need of a vehicle AEB simulation test system that can not only improve test efficiency and reduce costs, but also achieve dangerous extreme scene testing.

[0004] To summarize, the current vehicle AEB system testing scheme has problems such as large deviations in AEB system test results due to the difficulty in reproducing complex environments, high environmental reproduction costs, and long environmental reproduction cost cycles, which in turn leads to insufficient accuracy of vehicle safety inspection results, which is not conducive to driving safety of vehicle users. Summary of the invention

[0005] The main purpose of the present invention is to provide a testing method and system for a vehicle AEB system, so as to solve the problem that the vehicle AEB system testing scheme in the prior art has large deviations in AEB system test results due to difficulty in reproducing complex environments, high environmental reproduction costs and long environmental reproduction cost cycles, which in turn leads to insufficient accuracy of vehicle safety inspection results, which is not conducive to the driving safety of vehicle users.

[0006] In order to achieve the above object, according to one aspect of the present invention, a method for testing a vehicle AEB system is provided, comprising:

[0007] S1: generating a virtual test scene for a scene to be tested, wherein the virtual test scene includes a virtual vehicle representing a test vehicle;

[0008] S2: Collecting simulation video data in a virtual test scenario;

[0009] S3: setting the scene recognition accuracy of the virtual vehicle, and simulating the message information of the virtual vehicle in the virtual test scene according to the scene recognition accuracy, wherein the message information includes the whole vehicle message and the radar message;

[0010] S4: sending the simulation data including the simulation video data and the message information to the AEB system of the test vehicle, and outputting a control signal for the virtual vehicle, namely, a braking deceleration;

[0011] S5: convert the braking deceleration into a brake pedal signal;

[0012] S6: inputting the brake pedal signal into a dynamic model of the test vehicle, and outputting motion information of the test vehicle, wherein the motion information includes position information and posture information;

[0013] S7: updating the message information of the virtual vehicle in the virtual test scene according to the motion information;

[0014] S8: Compare the updated message information with the standard message information under the braking state, and output the test result of the AEB system.

[0015] Furthermore, S1 is specifically:

[0016] The virtual information of the virtual test scene is edited by scene simulation software to simulate the scene to be tested, wherein the virtual information includes roads, lane lines, paths, virtual vehicle conditions, traffic participating vehicle conditions, virtual vehicle driving road types, virtual vehicle motion information, traffic participating vehicle driving road types and traffic participating vehicle motion information. Lane lines include white solid lines, white dashed lines, yellow solid lines and yellow dashed lines; virtual vehicle driving road types include straight roads, curves, intersections, ramps and elevated roads; virtual vehicle motion information and traffic participating vehicle motion information both include the longitudinal position, lateral position and driving speed of the vehicle on the road to which it belongs.

[0017] Furthermore, after S1, it also includes:

[0018] Extract the difference information between the virtual test scene and the scene to be tested, and modify the virtual test scene according to the difference information.

[0019] Furthermore, the vehicle message includes AEB switch signal, vehicle speed signal, gear signal, steering wheel signal, accelerator pedal signal and brake pedal signal; the radar message includes the lateral relative distance, longitudinal relative distance, relative speed of the virtual vehicle relative to the target obstacle, as well as the target obstacle ID, dynamic attributes and detection percentage of the target obstacle.

[0020] Furthermore, the message information of the virtual vehicle is simulated in the virtual test scenario, specifically:

[0021] The message information of the virtual vehicle is simulated under the preset period and preset noise.

[0022] Furthermore, S4 specifically includes:

[0023] S401: Converting simulation data into vehicle-recognizable signals;

[0024] S402: Send the simulation data to the AEB system in the form of a vehicle-recognizable signal.

[0025] Further, the vehicle identifiable signal is a CAN signal; S401 is specifically:

[0026] The simulation data is converted into vehicle-recognizable signals through the CAN simulation device.

[0027] Furthermore, the brake pedal signal is a brake pedal opening signal; S5 is specifically:

[0028] The braking deceleration is converted into a brake pedal opening signal through the AB model.

[0029] Furthermore, the test results include qualified and unqualified; S8 is specifically:

[0030] When the updated message information conforms to the standard message information, the output test result is qualified; otherwise, the output test result is unqualified.

[0031] According to one aspect of the present invention, a test system for a vehicle AEB system is provided, comprising:

[0032] processor;

[0033] A memory having computer-readable instructions stored thereon, wherein when the computer-readable instructions are executed by the processor, the above-mentioned vehicle AEB system test is implemented.

[0034] By applying the technical solution of the present invention, the test vehicle and the test scene can be simulated in a virtual test scene, and an adjustable scene recognition accuracy of the virtual vehicle is set based on the test vehicle. Message information reflecting the whole vehicle information and radar information of the virtual vehicle in the virtual test scene is generated under the scene recognition accuracy. By generating message information with adjustable scene recognition accuracy, complex scenes can be flexibly simulated, test efficiency can be improved, and the cost of actual vehicle testing can be reduced, while the controllability of data and the ability to reproduce scenes can be improved. Then, the simulated video data and message information in the virtual test scene are transmitted to the AEB system in the form of recognizable data for calculation, and the control signal, i.e., the braking deceleration, is output, and converted into a brake pedal signal and substituted into the dynamic model of the test vehicle, and the motion information of the test vehicle is output, and then the message information of the virtual vehicle in the virtual test scene is updated according to the motion information, and the braking effect of the AEB system is judged according to the updated message information whether it meets the standard message information under the braking state, and the simulated video data and message information are transmitted to the AEB system to calculate and feedback the braking signal, which can realize the closed-loop verification of the virtual scene and the vehicle dynamics model, effectively evaluate the performance of the AEB system under different working conditions, improve the accuracy and automation level of the test, and avoid the high risk and high cost of the real vehicle test. Through this solution, complex real scenes can be reproduced in a virtual environment, avoiding the problem that the real scenes are difficult to reproduce, expanding the test completeness of the AEB system, improving the accuracy of the vehicle safety test results, and improving the driving safety of vehicle users. And there is no need to reproduce the real scene with a physical object, which greatly reduces the test cost of the AEB system.

[0035] In addition to the above-described purposes, features and advantages, the present invention has other purposes, features and advantages. The present invention will be further described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] The accompanying drawings constituting 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 an improper limitation of the present invention. In the accompanying drawings:

[0037] Figure 1 A schematic flow chart of a method for testing a vehicle AEB system provided by an embodiment of the present invention is shown;

[0038] Figure 2 A schematic diagram showing a comparison between simulated video data with noise and simulated video data without noise provided by an embodiment of the present invention is shown;

[0039] Figure 3 A data recording system of a virtual test scenario and a perception data recording system provided by an embodiment of the present invention is shown;

[0040] Figure 4 A schematic structural diagram of a vehicle AEB system test system provided by an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0041] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0042] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0043] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the terms used in this way can be interchanged where appropriate, so as to describe the embodiments of the present invention described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0044] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0045] Reference Manual Attached Figure 1 , shows a flow chart of a method for testing a vehicle AEB system provided by an embodiment of the present invention.

[0046] An embodiment of the present invention provides a method for testing a vehicle AEB system, including:

[0047] S1: Generate a virtual test scenario related to the scenario to be tested.

[0048] The virtual test scenario includes a virtual vehicle representing the test vehicle.

[0049] The test vehicle can be a commercial vehicle. A virtual test environment is created through the simulation system. The environment contains a virtual vehicle representing the actual test vehicle, which can simulate the road, vehicle and traffic conditions in the real scene, and provide basic scenario support for subsequent AEB system testing.

[0050] In a possible implementation manner, S1 specifically includes:

[0051] The virtual information of the virtual test scene is edited by scene simulation software to simulate the scene to be tested, wherein the virtual information includes roads, lane lines, paths, virtual vehicle conditions, traffic participating vehicle conditions, virtual vehicle driving road types, virtual vehicle motion information, traffic participating vehicle driving road types and traffic participating vehicle motion information.

[0052] It should be noted that the key virtual information of the virtual test scene is edited through the scene simulation software to fully simulate the actual situation of the scene to be tested. This virtual information includes road structure, lane markings, driving paths, as well as the dynamic conditions, motion characteristics and driving road types of virtual vehicles and traffic participating vehicles, ensuring that the virtual scene is highly consistent with the real test scene, providing an accurate simulation basis for subsequent tests.

[0053] In a possible implementation, lane lines include solid white lines, dashed white lines, solid yellow lines, and dashed yellow lines. The types of roads on which the virtual vehicle travels include straight roads, curves, intersections, ramps, and elevated roads. The motion information of the virtual vehicle and the motion information of the participating vehicles in the traffic include the longitudinal position, lateral position, and driving speed of the vehicle on the road to which it travels.

[0054] It is understandable that the lane line design of the virtual test scene covers common types such as white solid lines, dashed lines, yellow solid lines, and dashed lines to ensure the true restoration of the scene. At the same time, the types of roads that virtual vehicles travel on include straight roads, curves, intersections, ramps, and elevated roads, covering all different road environments. Vehicle motion information involves longitudinal position, lateral position, and driving speed, and describes in detail the dynamic performance of virtual vehicles and traffic participating vehicles on the road.

[0055] In a possible implementation manner, after S1, the method further includes:

[0056] Extract the difference information between the virtual test scene and the scene to be tested, and modify the virtual test scene according to the difference information.

[0057] It should be noted that after the initial editing of the virtual test scene is completed, the difference information between the virtual scene and the scene to be tested is extracted. The virtual scene is modified based on these differences to ensure that it is highly consistent with the actual test scene in terms of road characteristics, vehicle behavior and environmental dynamics, thereby improving the accuracy and reliability of the simulation test.

[0058] S2: Collect simulation video data in a virtual test scenario.

[0059] It is understandable that the video data of the virtual test scene is generated through simulation technology to truly reflect the dynamic interaction between the virtual vehicle and the surrounding environment, providing visualization and data support for subsequent AEB system testing. These video data can help verify the correctness of traffic elements in the virtual scene and the authenticity of dynamic performance, laying the foundation for the effectiveness of the test process.

[0060] S3: setting the scene recognition accuracy of the virtual vehicle, and simulating the message information of the virtual vehicle in a virtual test scene according to the scene recognition accuracy, wherein the message information includes the whole vehicle message and the radar message.

[0061] It should be noted that by setting the scene recognition accuracy of the virtual vehicle, simulating its operation under different recognition conditions, and generating vehicle messages and radar messages reflecting the vehicle status and environmental perception, the simulation process of these message information under preset conditions helps to verify the virtual vehicle's responsiveness to complex scenes and the reliability of sensor data.

[0062] In a possible implementation, the vehicle message includes an AEB switch signal, a vehicle speed signal, a gear position signal, a steering wheel signal, an accelerator pedal signal, and a brake pedal signal. The radar message includes the lateral relative distance, longitudinal relative distance, and relative speed of the virtual vehicle relative to the target obstacle, as well as the target obstacle ID, dynamic attributes, and detection percentage of the target obstacle.

[0063] Among them, dynamic attributes describe the dynamic characteristics of the target obstacle, such as whether the target is moving, the state of motion (uniform speed, acceleration, etc.), direction, etc. The detection percentage indicates the probability or confidence value of the target obstacle being successfully detected, usually provided in the form of a percentage, reflecting the reliability of the data. The vehicle message contains key information required for the operation of the AEB system, including the AEB switch status, vehicle speed, gear position, steering wheel angle, specific signals of the accelerator pedal and brake pedal, accurately describing the operating status of the vehicle. The radar message provides detailed perception information between the virtual vehicle and the target obstacle, which improves the test accuracy of the AEB system.

[0064] In a possible implementation, the message information of a virtual vehicle is simulated in a virtual test scenario, specifically:

[0065] The message information of the virtual vehicle is simulated under the preset period and preset noise.

[0066] It should be noted that those skilled in the art can set the preset period and the preset noise level according to actual needs, and the present invention is not limited thereto.

[0067] It should be noted that when simulating the message information of a virtual vehicle in a virtual test scenario, dynamic data that conforms to the actual situation is generated by setting a preset cycle and preset noise. The preset cycle determines the time interval for data updates, and the preset noise simulates signal fluctuations and errors in the real environment. This method flexibly allows technicians to adjust parameters according to test requirements, ensuring that the message information can reflect the real scene and meet the accuracy requirements of different test conditions, providing more realistic and reliable simulation support for subsequent tests.

[0068] Referring to Table 1, a whole vehicle message obtained by simulation is shown, and Table 1 gives each simulation message data and the purpose of each simulation message data.

[0069] Table 1 Vehicle message obtained by simulation

[0070] Signal use AEB function switch Determine whether the AEB function is turned on Speed ​​signal Determine the vehicle's driving status. Determine whether there is a risk of collision. Gear position signal Determine the vehicle's driving status Yaw rate signal Determine the vehicle's driving status Vehicle longitudinal acceleration signal Determine the vehicle's driving status Vehicle lateral acceleration signal Determine the vehicle's driving status Brake pedal signal Determining driver intentions Accelerator pedal position signal Determining driver intentions Turn signal Determining driver intentions Steering wheel angle signal Determine the vehicle's driving status and the driver's intention ESC activation signals Vehicle stability control system status

[0071] Table 2 Radar messages obtained by simulation

[0072]

[0073] S4: Sending the simulated data including the simulated video data and the message information to the AEB system of the test vehicle, and outputting a control signal for the virtual vehicle, namely, the braking deceleration.

[0074] Among them, the AEB system (Automatic Emergency Braking System) is an advanced driver assistance system. When the vehicle detects a potential collision ahead, the AEB will automatically trigger braking to reduce the risk of a collision or reduce the severity of a collision. The simulated video data and message information are passed as input to the AEB system of the test vehicle, allowing it to recognize the virtual test scene and generate a braking deceleration control signal. This signal reflects the response performance and braking decision of the AEB system in a specific virtual scene, providing basic data support for evaluating its safety.

[0075] In a possible implementation, S4 specifically includes:

[0076] S401: Convert the simulation data into a vehicle-recognizable signal.

[0077] In a possible implementation manner, the vehicle identifiable signal is a CAN signal. S401 is specifically:

[0078] The simulation data is converted into vehicle-recognizable signals through the CAN simulation device.

[0079] Among them, CAN simulation equipment refers to hardware or software tools that can convert simulation data into CAN (Controller Area Network) signals that can be recognized by vehicles, including but not limited to NI-CAN (National Instruments) and Ixxat CAN Interface.

[0080] The simulation data is converted into CAN signals that can be recognized by the vehicle through the CAN simulation device to achieve compatibility between the simulation data and the vehicle system. CAN signals are the standard form of internal communication in the vehicle. CAN simulation devices (such as NI-CAN or Ixxat CANInterface) convert simulation data into such signals, allowing the AEB system to accurately receive and process data from the virtual scene. In this way, it is ensured that the simulation data has actual usability in the vehicle system, providing effective input support for testing the response and performance of the AEB system.

[0081] S402: Send the simulation data to the AEB system in the form of a vehicle-recognizable signal.

[0082] It should be noted that the effective transmission of simulation data is achieved through two steps: first, the data generated by the simulation is converted into a signal form that can be recognized and processed by the vehicle, such as a CAN signal, to ensure that the data format is compatible with the AEB system. Then, the data is sent to the AEB system in this vehicle-recognizable signal form, so that it can complete scene perception and brake control calculations based on the input data. This method achieves effective docking between virtual scenes and real systems, providing a reliable data input channel for AEB performance testing.

[0083] S5: Converts the braking deceleration into a brake pedal signal.

[0084] It should be noted that the braking deceleration signal generated by the AEB system is converted into a brake pedal opening signal through a specific model to simulate the input behavior of the actual vehicle braking operation. This process enables the braking signal in the virtual scene to be closer to the braking control mechanism of the real vehicle, providing accurate braking input conditions for the subsequent calculation of the vehicle dynamics model.

[0085] In a possible implementation manner, the brake pedal signal is a brake pedal opening signal. S5 is specifically:

[0086] The braking deceleration is converted into a brake pedal opening signal through the AB model.

[0087] Among them, the AB model is a mathematical model used to describe the characteristics of the vehicle's braking system. It simulates the dynamic response of the braking system by associating the braking deceleration with the opening of the brake pedal. The AB model converts the braking deceleration generated by the AEB system into a brake pedal opening signal. Based on the physical and dynamic characteristics of the braking system, the model converts the vehicle's braking demand into the specific operation form of the brake pedal, thereby simulating the mechanical response of a real vehicle during braking. This method realizes the precise conversion of virtual signals to physical inputs, providing more realistic data support for subsequent vehicle dynamics calculations.

[0088] S6: Input the brake pedal signal into the dynamic model of the test vehicle, and output the motion information of the test vehicle, wherein the motion information includes position information and posture information.

[0089] Among them, the dynamic model is a mathematical model used to simulate the motion characteristics of the vehicle under various conditions. Based on the vehicle's mass, center of gravity position, tire characteristics, air resistance and other physical parameters and driving input (such as braking or acceleration signals), the vehicle's motion state in a specific scenario is calculated, including speed, acceleration, position and posture. By inputting the brake pedal signal into the dynamic model of the test vehicle, the actual motion information of the vehicle, including position and posture changes, is calculated. These data reflect the dynamic response of the vehicle under specific braking conditions, further verifying the impact of the AEB system on vehicle control and its performance.

[0090] S7: Update the message information of the virtual vehicle in the virtual test scene according to the motion information.

[0091] S8: Compare the updated message information with the standard message information under the braking state, and output the test result of the AEB system.

[0092] In a possible implementation manner, the test results include qualified and unqualified. S8 is specifically:

[0093] When the updated message information conforms to the standard message information, the output test result is qualified; otherwise, the output test result is unqualified.

[0094] It is understandable that the performance of the AEB system is evaluated by comparing the updated message information with the standard message information. If the message information accurately meets the standard requirements under braking, the test result is qualified. Otherwise, it is unqualified. This method ensures that the braking effect of the AEB system under different working conditions meets the safety and performance standards, providing a clear reference for system optimization and verification.

[0095] In actual application, the virtual simulation test process is used to verify the performance of the AEB system of the test vehicle. First, a virtual test environment that highly restores the scene to be tested is created through simulation software, and simulated video data is generated. In this environment, the scene recognition accuracy of the virtual vehicle is set, and message information reflecting the vehicle status and environmental perception is simulated. The simulation data is then input into the AEB system to trigger it to generate a control signal, namely the braking deceleration, and the deceleration is converted into a brake pedal signal. The signal is input into the dynamic model to calculate the vehicle's motion information, and then the message information in the virtual scene is updated based on the motion information. Finally, the updated message information is compared with the standard data to evaluate the braking effect of the AEB system. This closed-loop simulation test system effectively improves the efficiency, accuracy and safety of the test, while reducing the cost and risk of actual vehicle testing.

[0096] Refer to the instruction manual Figure 2 , shows a schematic diagram for comparing simulated video data with noise and simulated video data without noise provided by an embodiment of the present invention.

[0097] Figure 2 It is shown that the simulation system can output ideal video data as well as noisy video data. The ideal video data has no interference and can verify the early controller algorithm. The noisy video data can test the performance boundary of the controller algorithm.

[0098] Refer to the instruction manual Figure 3 , showing a data recording system of a virtual test scene and a perception data recording system provided by an embodiment of the present invention.

[0099] The virtual test scenario generated by the simulation system will be automatically saved to the perception data recording system after the test is completed. Since the virtual simulation test scenario is different from the real scenario, such as Figure 3 As shown. The data of the perception data recording system can be used for data feedback verification in the R&D department to accelerate the iteration speed of the controller algorithm. Since the virtual simulation test scene is generated by a high-performance graphics station and output through HDMI, the color, brightness, contrast, etc. may be different after being input into the controller. Therefore, the data storage function of the controller perception data recording system is opened, and R&D can feedback the data of the perception data recording system to the controller to verify the controller perception function indicators.

[0100] Specifically, a virtual simulation system can be built based on the scheme, including: a simulation system, a video transmission module, a real-time computing system, a CAN simulation device, an AB model, a perception data recording system, a bus data recording system, and a controller under test. The connection relationship between the various parts in the virtual simulation system is specifically as follows: the simulation system is connected to the real-time computing system: the simulation system provides virtual test scene data and receives feedback from the real-time computing system. The simulation system is connected to the video transmission module: the simulation video data generated by the simulation system is transmitted to the video transmission module. The video transmission module is connected to the controller under test: the video transmission module transmits the simulation video data to the controller under test for testing scene perception. The real-time computing system is connected to the controller under test through the CAN simulation device: the control signal generated by the real-time computing system is converted by the CAN simulation device and sent to the controller under test. The controller under test is connected to the perception data recording system: the perception and control data generated by the controller under test are transmitted to the perception data recording system for recording. The controller under test is connected to the bus data recording system: the control signal of the controller under test is simultaneously transmitted to the bus data recording system for recording. The bus data recording system is connected to the AB model: the braking signal in the bus data recording system is transmitted to the AB model for calculation of the braking opening. The AB model is connected to the real-time computing system: the brake opening signal calculated by the AB model is fed back to the real-time computing system for closed-loop optimization simulation. The above connections form a data flow closed loop between the modules for efficient simulation and testing of AEB system performance.

[0101] The simulation system is generally a simulation software that can realize the construction of virtual simulation scenes, vehicle dynamics model simulation, sensor simulation, and vehicle message simulation. The video transmission module is customized according to the video input protocol required by the controller. The video transmission module in this embodiment is an HDMI to MIPI protocol to realize the transmission of virtual simulation video data to the controller. The real-time computing system is generally a Linux real-time system, which communicates with the simulation system through Ethernet and communicates with the CAN simulation device through the bus protocol. It is mainly used to convert the simulation signal output by the simulation system into the real signal required by the controller, including CAN signal, LIN signal, Ethernet signal, and various analog digital signals. The CAN simulation device is mainly used to convert the signal simulated by the simulation system into a CAN signal and send it to the controller. The AB model is a self-built model, which is mainly used to realize different types of signal transmission. Generally, the controller outputs the vehicle control signal as braking deceleration or braking torque. Through the AB model, it can be converted into a brake pedal signal to achieve seamless connection with the vehicle dynamics model.

[0102] More specifically, first, the simulation system includes virtual test scenes, virtual sensors, and vehicle dynamics models. The simulation system generates virtual test scenes based on the road, lane lines, and vehicle motion information in real scenes, and injects the video stream data into the controller under test through the video transmission module to complete the simulation. Due to the signal path between the simulation system, the video transmission module, and the controller under test, the video screen noise and the recognition accuracy of the controller under test can be freely set through the simulation system to meet various test requirements. The video transmission module is used to complete the conversion from HDMI signal to MIPI signal. Secondly, the virtual test scene generated by the simulation system will be automatically saved to the perception data recording system after the test is completed. Since the virtual simulation test scene and the real data collected on the road are different in brightness, color, contrast, etc., the data of the perception data recording system can be used for data feedback verification in the R&D department to accelerate the iteration speed of the controller algorithm. Then, the simulation system completes the simulation of vehicle messages and radar messages. As a precondition for activating the AEB function of the controller, the simulation system performs simulation output according to the vehicle and radar technology protocols, and converts the simulation signal into a CAN signal and transmits it to the controller through the real-time computing system + CAN simulation equipment. The vehicle message includes but is not limited to AEB switch signal, vehicle speed signal, gear signal, steering wheel signal, accelerator pedal signal, brake pedal signal, etc., and the Raven message includes but is not limited to the horizontal and vertical relative distance of the target, relative speed, target ID, dynamic attributes, detection percentage, etc., and the cycle and noise of the simulation message can be dynamically adjusted, which is more flexible and operable than the real data collected on the road. Finally, after receiving the video data, radar data, and vehicle data, the controller under test activates the AEB function, outputs the vehicle control signal, and sends out the braking deceleration. The braking deceleration is converted into a brake pedal signal through the self-built AB model, and output to the vehicle dynamics model running in the real-time computing system. The vehicle dynamics model outputs its own position information and posture information after solving, and feeds back to the virtual test scene to complete the data closed loop of the AEB function test. In general, the bus data recording system is connected to the controller under test and the CAN simulation device data link, so that the message information during the vehicle AEB test can be recorded, including controller output messages and simulation system messages, and the vehicle controller function test verification can be completed based on standards and regulations.

[0103] For example, the simulation system is generally a scene simulation software, which runs in a high-performance map station and is used to render virtual simulation images. Some simulation systems provide vehicle dynamics model simulation functions. The construction of virtual scenes specifically includes: road editing, lane line editing, path selection, vehicle condition setting, and traffic participating vehicle condition setting. According to the test needs, select straight roads, curves, intersections, ramps, elevated roads and other road types, add lane lines on the road, such as white solid lines, white dashed lines, yellow solid lines, yellow dashed lines, etc., and then select a driving trajectory for the vehicle, set the longitudinal position, lateral position, and driving speed of the vehicle on the road, and set the longitudinal position, lateral position, and driving speed of the traffic participating vehicles on the road. The virtual simulation scene is output through HDMI, converted into a MIPI signal through the video transmission module, and sent to the controller with a delay of less than 10us. After being transmitted to the controller through the video transmission module, the perception and recognition distance accuracy is high. When the longitudinal recognition true value is 22.375m, the actual perception and recognition is 22.129m. When the true value of longitudinal recognition is 19.875m, the actual perceived recognition is 19.668m. When the true value of longitudinal recognition is 13.125m, the actual perceived recognition is 13.145m. When the true value of lateral recognition is 22.375m, the actual perceived recognition is 22.129m.

[0104] The virtual test scene generated by the simulation system will be automatically saved to the perception data recording system after the test is completed. Since the virtual simulation test scene is different from the real scene, the perception data recording system data can be used for data feedback verification in the R&D department to accelerate the iteration speed of the controller algorithm. Since the virtual simulation test scene is generated by a high-performance graphics station and output through HDMI, the color, brightness, contrast, etc. may be different after being input into the controller. Therefore, the data storage function of the controller perception data recording system is opened, and R&D can feedback the perception data recording system data to the controller to verify the controller perception function indicators.

[0105] After receiving the video data, radar data, and vehicle data, the controller under test activates the AEB function. The specific contents of the radar data and vehicle data are shown in Tables 1 and 2. After activating the AEB function, the controller outputs a vehicle control signal and sends it to the vehicle dynamics model running in the real-time system. The vehicle dynamics model outputs its own position information and posture information after solving and feeding it back to the virtual test scene, completing the data closed loop of the AEB function test. Specifically, after receiving the video data and radar data, the controller will first perform perception and recognition, including identifying the target ahead and lane line information. After the sensor data and vehicle data meet the AEB activation conditions, the controller will output the vehicle control command, i.e., the braking deceleration, according to the control logic of its own algorithm. The braking deceleration is converted into the brake pedal opening through the self-built AB model and input into the vehicle dynamics model. After receiving the deceleration signal, the dynamics model will decelerate, and then update its own motion information, and render it in the scene at the same time. At this time, the ideal sensor receives the updated data and feeds it back to the controller, completing the data closed loop AEB test.

[0106] Take the cut-in scenario as an example for specific explanation: the front vehicle (target vehicle) and the self-vehicle (test vehicle) are traveling in the same direction, and the self-vehicle and the front vehicle are driving in two adjacent lanes respectively. The self-vehicle maintains a speed of 50km / h and drives in a straight line, the target vehicle drives at a speed of 20km / h, and the front vehicle cuts in front of the self-vehicle. When the cut-in is completed, the longitudinal distance between the two vehicles is 20m. When the two vehicles are driving at a constant speed, the self-vehicle is equipped with an ideal camera and radar sensor. The ideal camera is responsible for inputting video data to the controller through the HDMI level 1 video transmission module. The ideal radar obtains the target information in front and inputs it to the controller through the real-time system and CAN card. When TTC (the relative distance between the two vehicles divided by the relative speed) meets the controller AEB activation condition, the controller AEB function will be activated. Specifically, the controller obtains the relative speed and relative distance of the target in front through the sensor fusion method of the camera and radar. When the two vehicles are driving at a constant speed, the camera recognizes the vehicle in front, but does not regard it as a CIPV (Closest In-Path Vehicle). When the vehicle in front has a tendency to cut into the lane of the self-vehicle, it is recognized as a CIPV (Closest In-Path Vehicle) by the camera. When the longitudinal distance between the two vehicles is 38.293m, it is regarded as a CIPV (Closest In-Path Vehicle). When the longitudinal distance between the two vehicles is 23.488m, the controller's first-level warning is activated, and the controller issues a sound reminder signal. When the longitudinal distance between the two vehicles is 19.941m, the controller's second-level warning is activated, and the controller issues a smaller braking deceleration. When the longitudinal distance between the two vehicles is 13.223m, the controller's third-level warning is activated, and the controller issues the maximum braking deceleration to avoid collision between the self-vehicle and the vehicle in front.

[0107] The vehicle dynamics model receives the brake pedal opening signal (between 0-1, 0 represents 0% opening, 1 represents 100% opening), and needs to convert the brake deceleration into the brake pedal opening. The signal conversion is realized by the self-built model. The specific implementation logic of the signal conversion model is: the model receives the accelerator pedal opening and brake pedal opening of the simulation system driver model. When AEB is activated, both openings are greater than or equal to 0. When accelerating, the accelerator pedal is greater than 0 and the brake pedal is equal to 0. When decelerating, the accelerator pedal is equal to 0 and the brake pedal is greater than 0. After the controller sends out the brake deceleration, it is subtracted from the actual deceleration of the vehicle and the difference is input into the signal conversion model. The signal conversion model outputs the brake pedal opening according to its own logic, so that the vehicle slows down or stops. When the vehicle speed is 50km / h, the relationship between the brake pedal opening output by the signal conversion model and the real vehicle will be compared and analyzed, so that the braking performance of the simulated vehicle at 50km / h is consistent with the braking performance of the real vehicle at 50km / h. Due to the change in the speed of the vehicle, the virtual scene is updated, and the closed-loop verification of the controller AEB function is completed.

[0108] The bus data recording system is connected to the AEB system of the controller under test and the CAN simulation device data link, and the message information during the vehicle AEB test can be recorded, including the controller output message and the simulation system message, and the performance evaluation of the controller under test can be completed based on standards and regulations. Specifically, the relative speed and relative distance of the front vehicle identified during the perception process, and the time and distance of the front target as the main target are recorded.

[0109] It is understandable that by generating virtual simulation test scenarios based on the simulation system, a large number of scenarios can be reproduced. This includes standard regulatory scenarios, dangerous extreme scenarios, and real physical world scenario reproduction. It is also possible to generalize parameters based on a real scenario, generate test scenarios in batches, and conduct comprehensive testing on the controller, with high test efficiency and low test cost. The conversion of simulated video data to real video data is realized through the video transmission module, and the perception and recognition error can be less than 1%, which greatly improves the accuracy of the video dark box and saves a lot of calibration work compared to the actual vehicle test. The actual vehicle simulation is completed based on the commercial vehicle dynamics model of the simulation system and the actual vehicle design parameters, which greatly speeds up the test efficiency and can verify the performance of different models under the same controller. The signals required by the controller, including video data, radar data, and vehicle data, are all simulated through the simulation system, and the parameters such as data accuracy and cycle can be adjusted to meet the different test requirements of the controller. Multi-source data accumulation includes simulation scene data, perception data recording system video data, and bus test result data, all of which are saved locally. Simulation scene data can be used as the source of different controller test data, perception data recording system data can verify the controller perception function, and bus test result data can fully record the entire test process and key data information, making the test process completely traceable. At the same time, by analyzing the test result data, the controller function evaluation can be completed according to standards and regulations. This system is not limited to controller interfaces or vehicle models, and truly completes the commercial vehicle AEB function verification with one system, reducing R&D testing costs.

[0110] In actual application, based on the virtual test scene generated by the simulation system, the video data is transmitted to the controller under test through the video transmission module for perception and recognition. The perception accuracy and error can be adjusted by the simulation system. During the perception and recognition process, the data is automatically saved in the perception data recording system to facilitate data re-injection for research and development. The video data in the perception data recording system and the video data generated by the simulation system are different in color, brightness, and contrast. At the same time, the simulation of the vehicle message and radar message required for the activation of the function of the controller under test is completed in the simulation system, and the virtual signal is converted into a CAN signal through the real-time computing system with the CAN simulation device and sent to the controller under test. Since both the video data and the bus data are generated by the simulation system, data transmission is achieved synchronously. After receiving video data, radar data, and vehicle data, the controller under test activates the AEB function after satisfying the controller algorithm logic, outputs the vehicle control signal, and sends it to the commercial vehicle dynamics model running in the real-time computing system after AB model conversion. The commercial vehicle dynamics model simulates the real vehicle by importing the real vehicle parameters into the virtual vehicle model. The internal solving unit of the commercial vehicle dynamics model calculates and outputs the position information and posture information of the simulated vehicle in the virtual test scene. The virtual sensor in the simulation system detects the updated scene information to complete the data closed loop of the AEB function test. Through the bus data recording system, the message information in the data closed loop process can be saved, and the test report of the controller under test can be output based on the controller function specifications and standard regulations to evaluate the controller performance indicators.

[0111] From the above description, it can be seen that by applying the technical solution of the present invention, the test vehicle and the test scene can be simulated in the virtual test scene, and an adjustable scene recognition accuracy of the virtual vehicle is set based on the test vehicle. Message information reflecting the whole vehicle information and radar information of the virtual vehicle in the virtual test scene is generated under this scene recognition accuracy. By generating message information with adjustable scene recognition accuracy, complex scenes can be flexibly simulated, test efficiency can be improved, and the cost of actual vehicle testing can be reduced, while the controllability of data and the ability to reproduce scenes can be improved. Then, the simulated video data and message information in the virtual test scene are transmitted to the AEB system in the form of recognizable data for calculation, and the control signal, i.e., the braking deceleration, is output, and converted into a brake pedal signal and substituted into the dynamic model of the test vehicle, and the motion information of the test vehicle is output, and then the message information of the virtual vehicle in the virtual test scene is updated according to the motion information, and the braking effect of the AEB system is judged according to the updated message information whether it meets the standard message information under the braking state, and the simulated video data and message information are transmitted to the AEB system to calculate and feedback the braking signal, which can realize the closed-loop verification of the virtual scene and the vehicle dynamics model, effectively evaluate the performance of the AEB system under different working conditions, improve the accuracy and automation level of the test, and avoid the high risk and high cost of the real vehicle test. Through this solution, complex real scenes can be reproduced in a virtual environment, avoiding the problem that the real scenes are difficult to reproduce, expanding the test completeness of the AEB system, improving the accuracy of the vehicle safety test results, and improving the driving safety of vehicle users. And there is no need to reproduce the real scene with a physical object, which greatly reduces the test cost of the AEB system.

[0112] Reference Manual Attached Figure 4 , showing a schematic structural diagram of a vehicle AEB system testing system provided by an embodiment of the present invention.

[0113] The embodiment of the present invention provides a vehicle AEB system testing system 20, comprising:

[0114] Processor 201;

[0115] The memory 202 stores computer-readable instructions, and when the computer-readable instructions are executed by the processor 201 , the above-mentioned vehicle AEB system testing method is implemented.

[0116] From the above description, it can be seen that by applying the technical solution of the present invention, the test vehicle and the test scene can be simulated in the virtual test scene, and an adjustable scene recognition accuracy of the virtual vehicle is set based on the test vehicle. Message information reflecting the whole vehicle information and radar information of the virtual vehicle in the virtual test scene is generated under this scene recognition accuracy. By generating message information with adjustable scene recognition accuracy, complex scenes can be flexibly simulated, test efficiency can be improved, and the cost of actual vehicle testing can be reduced, while the controllability of data and the ability to reproduce scenes can be improved. Then, the simulated video data and message information in the virtual test scene are transmitted to the AEB system in the form of recognizable data for calculation, and the control signal, i.e., the braking deceleration, is output, and converted into a brake pedal signal and substituted into the dynamic model of the test vehicle, and the motion information of the test vehicle is output, and then the message information of the virtual vehicle in the virtual test scene is updated according to the motion information, and the braking effect of the AEB system is judged according to the updated message information whether it meets the standard message information under the braking state, and the simulated video data and message information are transmitted to the AEB system to calculate and feedback the braking signal, which can realize the closed-loop verification of the virtual scene and the vehicle dynamics model, effectively evaluate the performance of the AEB system under different working conditions, improve the accuracy and automation level of the test, and avoid the high risk and high cost of the real vehicle test. Through this solution, complex real scenes can be reproduced in a virtual environment, avoiding the problem that the real scenes are difficult to reproduce, expanding the test completeness of the AEB system, improving the accuracy of the vehicle safety test results, and improving the driving safety of vehicle users. And there is no need to reproduce the real scene with a physical object, which greatly reduces the test cost of the AEB system.

[0117] Unless otherwise specifically stated, the relative arrangement of the parts and steps described in these embodiments, numerical expressions and numerical values ​​do not limit the scope of the present invention. At the same time, it should be understood that, for ease of description, the sizes of the various parts shown in the accompanying drawings are not drawn according to the actual proportional relationship. The technology, method and equipment known to ordinary technicians in the relevant field may not be discussed in detail, but in appropriate cases, the technology, method and equipment should be regarded as a part of the authorization specification. In all examples shown and discussed here, any specific value should be interpreted as being merely exemplary, rather than as a limitation. Therefore, other examples of exemplary embodiments may have different values. It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once a certain item is defined in an accompanying drawing, it does not need to be further discussed in subsequent drawings.

[0118] For ease of description, spatially relative terms such as "above", "above", "on the upper surface of", "above", etc. may be used here to describe the spatial positional relationship between a device or feature and other devices or features as shown in the figure. It should be understood that spatially relative terms are intended to include different orientations of the device in use or operation in addition to the orientation described in the figure. For example, if the device in the accompanying drawings is inverted, the device described as "above other devices or structures" or "above other devices or structures" will be positioned as "below other devices or structures" or "below other devices or structures". Thus, the exemplary term "above" can include both "above" and "below". The device can also be positioned in other different ways (rotated 90 degrees or in other orientations), and the spatially relative descriptions used here are interpreted accordingly.

[0119] In the description of the present invention, it is necessary to understand that the directions or positional relationships indicated by directional words such as "front, back, up, down, left, right", "lateral, vertical, perpendicular, horizontal" and "top, bottom" are usually based on the directions or positional relationships shown in the drawings. They are only for the convenience of describing the present invention and simplifying the description. Unless otherwise specified, these directional words do not indicate or imply that the devices or elements referred to must have a specific direction or be constructed and operated in a specific direction. Therefore, they cannot be understood as limiting the scope of protection of the present invention. The directional words "inside and outside" refer to the inside and outside relative to the contours of each component itself.

[0120] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for testing a vehicle AEB system, characterized in that: include: S1: generating a virtual test scene related to a scene to be tested, wherein the virtual test scene includes a virtual vehicle representing a test vehicle; S2: Collecting simulation video data in the virtual test scene; S3: setting the scene recognition accuracy of the virtual vehicle, and simulating the message information of the virtual vehicle in the virtual test scene according to the scene recognition accuracy, wherein the message information includes the whole vehicle message and the radar message; S4: sending the simulation data including the simulation video data and the message information to the AEB system of the test vehicle, and outputting a control signal for the virtual vehicle, namely, a braking deceleration; S5: converting the braking deceleration into a brake pedal signal; S6: inputting the brake pedal signal into the dynamic model of the test vehicle, and outputting motion information of the test vehicle, wherein the motion information includes position information and posture information; S7: updating the message information of the virtual vehicle in the virtual test scene according to the motion information; S8: Compare the updated message information with the standard message information under the braking state, and output the test result of the AEB system.

2. The vehicle AEB system testing method according to claim 1, characterized in that: The S1 is specifically: The virtual information of the virtual test scene is edited by scene simulation software to simulate the scene to be tested, wherein the virtual information includes roads, lane lines, paths, virtual vehicle conditions, traffic participating vehicle conditions, virtual vehicle driving road types, virtual vehicle motion information, traffic participating vehicle driving road types and traffic participating vehicle motion information, the lane lines include white solid lines, white dotted lines, yellow solid lines and yellow dotted lines; the virtual vehicle driving road types include straight roads, curves, intersections, ramps and elevated roads; the virtual vehicle motion information and the traffic participating vehicle motion information both include the longitudinal position, lateral position and driving speed of the vehicle on the corresponding driving road.

3. The vehicle AEB system testing method according to claim 1, characterized in that: After S1, the following is also included: Extracting difference information between the virtual test scene and the scene to be tested, and correcting the virtual test scene according to the difference information.

4. The vehicle AEB system testing method according to claim 1, characterized in that: The whole vehicle message includes an AEB switch signal, a vehicle speed signal, a gear signal, a steering wheel signal, an accelerator pedal signal and a brake pedal signal; the radar message includes a lateral relative distance, a longitudinal relative distance, a relative speed of the virtual vehicle relative to the target obstacle, and a target obstacle ID, dynamic attributes and detection percentage of the target obstacle.

5. The vehicle AEB system testing method according to claim 1, characterized in that: The simulating of the message information of the virtual vehicle in the virtual test scenario is specifically as follows: The message information of the virtual vehicle is simulated under a preset period and preset noise.

6. The vehicle AEB system testing method according to claim 1, characterized in that: The S4 specifically includes: S401: Converting the simulation data into a vehicle-recognizable signal; S402: Sending the simulation data to the AEB system in the form of a signal recognizable by the vehicle.

7. The vehicle AEB system testing method according to claim 6, characterized in that: The vehicle identifiable signal is a CAN signal; S401 is specifically: The simulation data is converted into a signal recognizable by the vehicle through a CAN simulation device.

8. The vehicle AEB system testing method according to claim 1, characterized in that: The brake pedal signal is a brake pedal opening signal; S5 is specifically: The braking deceleration is converted into a brake pedal opening signal through the AB model.

9. The vehicle AEB system testing method according to claim 1, characterized in that: The test results include qualified and unqualified; S8 specifically includes: In the case where the updated message information conforms to the standard message information, the test result is output as qualified; otherwise, the test result is output as unqualified.

10. A vehicle AEB system testing system, characterized in that: include: processor; A memory having computer-readable instructions stored thereon, wherein when the computer-readable instructions are executed by the processor, the vehicle AEB system testing method according to any one of claims 1 to 9 is implemented.

Images