Vehicle controller test method and device and electronic device

Abstract

The invention discloses a vehicle controller testing method and device and an electronic device, and relates to the technical field of automatic driving. The method comprises the steps that first pose information of a vehicle is acquired, and the vehicle is a vehicle running on a real test road; virtual environment information is injected into multiple sensors of the vehicle, fault signals are injected into the multiple sensors at the target moment, the virtual vehicle is used for simulating the driving state of the vehicle in the virtual scene, and the target moment is the moment when the virtual obstacle executes the dangerous behavior; controlling a vehicle controller to generate an execution decision according to the sensing information of the plurality of sensors after the fault signals are injected and the first pose information; it is determined whether the virtual vehicle safely passes the virtual obstacle based on the execution decision to test the vehicle controller. According to the invention, the technical problem that the stability of the vehicle is insufficient when only a single sensor is tested in the prior art is solved.

Description

Technical Field

[0001] The present invention relates to the field of autonomous driving technology, and in particular to a vehicle controller testing method, device, and electronic device. Background Art

[0002] With the development of intelligent driving technology, multi-sensor fusion perception has become mainstream in autonomous vehicles. Through the coordinated operation of sensors such as cameras, radar, and lidar, vehicles can perceive their surroundings, thereby achieving autonomous driving. However, existing technologies for fault scenario testing primarily focus on simulating single sensor failures, which fails to fully verify the stability and robustness of the vehicle under fault conditions.

[0003] No effective solution has been proposed to the above problems. Summary of the Invention

[0004] Embodiments of the present invention provide a vehicle controller testing method, device, and electronic device to at least solve the technical problem of insufficient vehicle stability when only a single sensor is tested in the related art.

[0005] According to one embodiment of the present invention, a vehicle controller testing method is provided, comprising: obtaining first-position posture information of a vehicle, wherein the vehicle is a vehicle traveling on a real test road; injecting virtual environment information into multiple sensors of the vehicle, and injecting fault signals into the multiple sensors at a target moment, wherein the virtual environment information is environmental information of a virtual scene, the virtual scene is used to simulate a vehicle controller test scene, the virtual scene includes a virtual obstacle and a virtual vehicle, the virtual vehicle is used to simulate the driving state of the vehicle in the virtual scene, and the target moment is the moment when the virtual obstacle performs a dangerous behavior; controlling the vehicle controller to generate an execution decision based on the perception information and first-position information of the multiple sensors after the fault signal is injected, wherein the perception information is data collected from the virtual scene by the multiple sensors after the fault signal is injected; and determining whether the virtual vehicle safely passes through the virtual obstacle based on the execution decision to test the vehicle controller.

[0006] Optionally, before injecting fault signals into the multiple sensors at the target time, the vehicle controller testing method also includes: dynamically adjusting the virtual posture of the virtual vehicle based on the first posture information to obtain the first virtual vehicle posture information of the virtual vehicle; and controlling the virtual vehicle to travel in the virtual scene based on the virtual environment information and the first virtual vehicle posture information.

[0007] Optionally, the vehicle controller is controlled to generate an execution decision based on the perception information of multiple sensors after the fault signal is injected and the first posture information, including: controlling the vehicle controller to generate an execution decision based on the perception information of multiple sensors after the fault signal is injected and the first posture information based on the decision generation algorithm; or, obtaining the driver's execution operation, wherein the execution operation is the driver's operation of avoiding the virtual obstacle in combination with the display information of the target screen, and the target screen is used to display the virtual scene; controlling the vehicle controller to generate an execution decision based on the decision generation algorithm according to the execution operation, the perception information of multiple sensors after the fault signal is injected and the first posture information.

[0008] Optionally, determining whether the virtual vehicle has safely passed through the virtual obstacle based on the execution decision includes: controlling the vehicle to perform a preset operation based on the execution decision to obtain second posture information of the vehicle, wherein the preset operation is used to avoid the virtual obstacle; dynamically adjusting the virtual posture of the virtual vehicle based on the second posture information to obtain second virtual vehicle posture information of the virtual vehicle; determining the virtual obstacle posture information of the virtual obstacle based on the second posture information and the perception information; determining whether the virtual vehicle has safely passed through the virtual obstacle based on the second virtual vehicle posture information and the virtual obstacle posture information to obtain a determination result.

[0009] Optionally, the vehicle controller testing method further includes: in response to a determination result indicating that the virtual vehicle safely passes through the virtual obstacle, determining that the vehicle controller test has passed; in response to a determination result indicating that the virtual vehicle does not safely pass through the virtual obstacle, determining that the vehicle controller test has failed, and updating the decision generation algorithm.

[0010] Optionally, the vehicle controller testing method further includes: in response to a failure of a first sensor among multiple sensors, reducing the weight of the first sensor and increasing the weight of a second sensor, wherein the first sensor is part of the multiple sensors and the second sensor is a sensor other than the first sensor among the multiple sensors; determining perception information based on the multiple sensors after weight adjustment; or in response to a failure of multiple sensors, fusing the perception data of the multiple sensors to obtain fused perception data; and determining the perception information based on the fused perception data.

[0011] Optionally, obtaining the first position information of the vehicle includes: determining the position information and the position information of the vehicle based on the navigation system and the positioning system of the vehicle to obtain the first position information.

[0012] According to one embodiment of the present invention, a vehicle controller testing device is also provided, the device comprising: an acquisition module, the acquisition module being used to acquire the first posture information of the vehicle, wherein the vehicle is a vehicle traveling on a real test road; an injection module, the injection module being used to inject virtual environment information into multiple sensors of the vehicle, and injecting fault signals into the multiple sensors at a target time, wherein the virtual environment information is the environmental information of a virtual scene, the virtual scene is used to simulate a vehicle controller test scene, the virtual scene includes a virtual obstacle and a virtual vehicle, the virtual vehicle is used to simulate the driving state of the vehicle in the virtual scene, and the target time is the time when the virtual obstacle performs a dangerous behavior; a control module, the control module being used to control the vehicle controller to generate an execution decision based on the perception information and the first posture information of the multiple sensors after the fault signal is injected, wherein the perception information is the data collected from the virtual scene by the multiple sensors after the fault signal is injected; a testing module, the testing module being used to determine whether the virtual vehicle safely passes through the virtual obstacle based on the execution decision, so as to test the vehicle controller.

[0013] Optionally, before injecting fault signals into the multiple sensors at the target time, the vehicle controller testing device also includes an adjustment module for dynamically adjusting the virtual posture of the virtual vehicle based on the first posture information to obtain the first virtual vehicle posture information of the virtual vehicle; and controlling the virtual vehicle to travel in the virtual scene based on the virtual environment information and the first virtual vehicle posture information.

[0014] Optionally, the control module is also used to control the vehicle controller to generate an execution decision based on a decision generation algorithm according to the perception information of multiple sensors after the fault signal is injected and the first posture information; or, obtain the driver's execution operation, wherein the execution operation is the driver's operation of avoiding the virtual obstacle in combination with the display information of the target screen, and the target screen is used to display the virtual scene; control the vehicle controller to generate an execution decision based on the decision generation algorithm according to the execution operation, the perception information of multiple sensors after the fault signal is injected and the first posture information.

[0015] Optionally, the test module is also used to control the vehicle to perform a preset operation based on the execution decision to obtain second posture information of the vehicle, wherein the preset operation is used to avoid a virtual obstacle; dynamically adjust the virtual posture of the virtual vehicle based on the second posture information to obtain second virtual vehicle posture information of the virtual vehicle; determine the virtual obstacle posture information of the virtual obstacle based on the second posture information and the perception information; determine whether the virtual vehicle has safely passed through the virtual obstacle based on the second virtual vehicle posture information and the virtual obstacle posture information to obtain a determination result.

[0016] Optionally, the vehicle controller testing device also includes a first determination module for determining that the vehicle controller test has passed in response to a determination result indicating that the virtual vehicle has safely passed the virtual obstacle; and for determining that the vehicle controller test has failed in response to a determination result indicating that the virtual vehicle has not safely passed the virtual obstacle, and updating the decision generation algorithm.

[0017] Optionally, the vehicle controller testing device also includes a second determination module for reducing the weight of the first sensor and increasing the weight of the second sensor in response to a failure of the first sensor among multiple sensors, wherein the first sensor is part of the multiple sensors and the second sensor is a sensor other than the first sensor among the multiple sensors; determining perception information based on the multiple sensors after weight adjustment; or in response to a failure of multiple sensors, fusing the perception data of the multiple sensors to obtain fused perception data; and determining perception information based on the fused perception data.

[0018] Optionally, the acquisition module is further used to determine the position information and posture information of the vehicle based on the vehicle's navigation system and positioning system to obtain the first position information.

[0019] According to one embodiment of the present invention, a vehicle is provided, which is used to execute any of the above vehicle controller testing methods.

[0020] According to one embodiment of the present invention, a computer-readable storage medium is further provided, in which a computer program is stored. The computer program is configured to execute any of the above-mentioned vehicle controller testing methods when running on a computer or a processor.

[0021] According to one embodiment of the present invention, an electronic device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to run the computer program to execute any of the above vehicle controller testing methods.

[0022] According to one embodiment of the present invention, a computer program product is further provided, comprising a computer program, which implements any of the above vehicle controller testing methods when executed by a processor.

[0023] In an embodiment of the present invention, by obtaining the first posture information of a vehicle, wherein the vehicle is a vehicle traveling on a real test road; injecting virtual environment information into multiple sensors of the vehicle, and injecting fault signals into the multiple sensors at a target time, wherein the virtual environment information is environmental information of a virtual scene, the virtual scene is used to simulate a vehicle controller test scene, the virtual scene includes a virtual obstacle and a virtual vehicle, the virtual vehicle is used to simulate the driving state of the vehicle in the virtual scene, and the target time is the time when the virtual obstacle performs a dangerous behavior; controlling the vehicle controller to generate an execution decision based on the perception information and the first posture information of the multiple sensors after the fault signal is injected, wherein the perception information is data collected by the multiple sensors from the virtual scene after the fault signal is injected; determining whether the virtual vehicle safely passes through the virtual obstacle based on the execution decision to test the vehicle controller, thereby achieving the purpose of comprehensively testing the fault response capability of the vehicle controller under the collaborative work of multiple sensors, thereby achieving the technical effect of improving the safety and reliability of the autonomous driving vehicle in complex fault scenarios, and thus solving the technical problem of insufficient vehicle stability when only a single sensor is tested in the related art. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. 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 drawings:

[0025] Figure 1 is a flow chart of a vehicle controller testing method according to one embodiment of the present invention;

[0026] Figure 2 is a diagram of a fault injection test scenario according to one embodiment of the present invention;

[0027] Figure 3 is a structural diagram of a vehicle-in-the-loop system according to one embodiment of the present invention;

[0028] Figure 4 FIG. 4 is a structural block diagram of a vehicle controller testing device according to one embodiment of the present invention. DETAILED DESCRIPTION

[0029] To facilitate understanding, some descriptions of concepts related to the embodiments of the present invention are exemplarily provided for reference.

[0030] As shown below:

[0031] Vehicle-in-the-Loop (VIL) is a simulation testing technology that integrates a real vehicle or its subsystems into a virtual test environment. In VIL testing, the vehicle's real actuators (such as the engine, suspension, and braking system) interact with other vehicles, pedestrians, road conditions, and weather factors in the virtual environment.

[0032] Sensor fusion refers to the process of integrating the output data of multiple sensors (such as cameras, radars, and lidars) to improve the system's perception and decision-making capabilities.

[0033] Fault injection is a system testing technique designed to evaluate system behavior by artificially introducing faults or abnormal conditions. In autonomous driving system development, fault injection is often used to simulate sensor failures, network delays, power supply fluctuations, and other conditions to test the system's robustness and fault tolerance under these abnormal conditions.

[0034] Multi-sensor collaboration refers to two or more sensors working together to provide more complete and accurate information than a single sensor could provide. In autonomous driving technology, multi-sensor collaboration is crucial because different sensors have different advantages and limitations in different environmental conditions.

[0035] The Kalman filter is a recursive mathematical algorithm used to estimate the state of a system. In the presence of measurement noise, the Kalman filter can effectively combine the measurements of multiple sensors to minimize the estimation error.

[0036] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions 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 embodiments described 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 making creative efforts should fall within the scope of protection of the present invention.

[0037] It should be noted that the terms "first", "second", etc. in the description 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 data used in this way can be interchangeable where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In the description of this embodiment, unless otherwise specified, the meaning of "plurality" is two or more. 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 clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0038] According to one embodiment of the present invention, an embodiment of a vehicle controller testing method is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0039] The method embodiment can be executed in an electronic device, a similar control device, or a system including a memory and a processor. Taking an electronic device as an example, the electronic device may include one or more processors and a memory for storing data. Optionally, the electronic device may also include a communication device and a display device for communication functions. It will be understood by those skilled in the art that the above structural description is only illustrative and does not limit the structure of the electronic device. For example, the electronic device may also include more or fewer components than those described in the above structural description, or have a configuration different from the above structural description.

[0040] The processor may include one or more processing units. For example, the processor may include a central processing unit (CPU), a graphics processing unit (GPU), a digital signal processing (DSP) chip, a microcontroller unit (MCU), a programmable logic device (field-programmable gate array, FPGA), a neural network processor (NPU), a tensor processing unit (TPU), an artificial intelligence (AI) type processor, and the like. Among them, different processing units may be independent components or integrated into one or more processors. In some instances, the electronic device may also include one or more processors.

[0041] The memory can be used to store computer programs, such as a computer program corresponding to the vehicle controller testing method in an embodiment of the present invention. The processor implements the above-mentioned vehicle controller testing method by running the computer program stored in the memory. The memory may include a high-speed random access memory and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory may further include a memory remotely located relative to the processor, and these remote memories may be connected to the electronic device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0042] The communication device is used to receive or send data via a network. Specific examples of the aforementioned network may include a wireless network provided by a communication provider of the mobile terminal. In one embodiment, the communication device includes a network interface controller (NIC), which can be connected to other network devices via a base station to enable communication with the Internet. In one embodiment, the communication device may be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0043] The display device can be, for example, a touch screen liquid crystal display (LCD) and a touch display (also referred to as a "touch screen" or "touch display screen"). The liquid crystal display can enable the user to interact with the user interface of the mobile terminal. In some embodiments, the mobile terminal has a graphical user interface (GUI), and the user can interact with the GUI by finger contacts and / or gestures on the touch-sensitive surface. The human-computer interaction functions here optionally include the following interactions: creating web pages, drawing, word processing, making electronic documents, games, video conferencing, instant messaging, sending and receiving emails, call interfaces, playing digital videos, playing digital music and / or web browsing, etc. The executable instructions for performing the above-mentioned human-computer interaction functions are configured / stored in a computer program product or readable storage medium executable by one or more processors.

[0044] In this embodiment, a vehicle controller testing method running on an electronic device is provided. Figure 1 FIG. 1 is a flow chart of a vehicle controller testing method according to one embodiment of the present invention. Figure 1 As shown, the process includes the following steps:

[0045] Step S10, obtaining first position information of the vehicle, wherein the vehicle is a vehicle traveling on a real test road;

[0046] In the embodiment of the present invention, the first position information may be understood as the initial position, direction, and posture information of the vehicle at the beginning of the test.

[0047] A real test road can be understood as the actual road surface used to test a vehicle's autonomous driving system. This can range from city streets to highways, or it can be a specialized test track designed specifically for vehicle testing. Real test roads may have specific test conditions pre-set, such as varying road conditions, weather conditions, and traffic density, which are not restricted here.

[0048] Obtaining the vehicle's first pose information can be understood as extracting detailed information about the vehicle's position, orientation, and attitude at a given moment from the vehicle's positioning and attitude system. For example, the vehicle's position information, heading (yaw), pitch (pitch), and roll (roll) information, as well as the vehicle's speed and acceleration are obtained, though this is not a limitation.

[0049] In the embodiment of the present invention, by obtaining accurate posture information of a vehicle traveling on a real test road, including position, direction, and motion state, a data basis is provided for subsequent tests, thereby enhancing the authenticity and effectiveness of the test.

[0050] Step S12: injecting virtual environment information into multiple sensors of the vehicle and injecting fault signals into the multiple sensors at a target time, wherein the virtual environment information is environmental information of a virtual scene, the virtual scene is used to simulate a vehicle controller test scene, the virtual scene includes a virtual obstacle and a virtual vehicle, the virtual vehicle is used to simulate the driving state of the vehicle in the virtual scene, and the target time is the time when the virtual obstacle performs a dangerous behavior;

[0051] In an embodiment of the present invention, multiple sensors can be understood as various sensing devices installed on the vehicle, including but not limited to forward-looking cameras, millimeter-wave radars, laser radars (LiDAR), etc., which are used to collect data on the vehicle's surrounding environment, such as images, obstacle distances, speed, etc., which are not limited here.

[0052] Virtual environment information can be understood as environmental data in a virtual scene, including road layout, traffic signs, pedestrians, weather conditions, etc., which are not limited here.

[0053] The target moment can be understood as a specific time point preset in the virtual scene. At this time, the virtual obstacle will perform dangerous behaviors, such as a pedestrian suddenly appearing in front of the vehicle (ghosting phenomenon) or a vehicle in the adjacent lane making an emergency turn. There are no restrictions here.

[0054] The fault signal can be understood as sending simulated error or abnormal data to the vehicle's sensors through the fault injection module at the target moment to simulate sensor failures that may be encountered in the real world, such as camera blur and radar signal loss.

[0055] A virtual scene can be understood as a three-dimensional model created by simulation software, which simulates various real-world driving scenarios that a vehicle may encounter, including but not limited to urban roads, highways, country roads, etc., as well as various static and dynamic obstacles.

[0056] Virtual obstacles can be understood as obstacles artificially set up in a virtual scene. They can be static, such as roadblocks and parked vehicles, or dynamic, such as pedestrians and other moving vehicles. There is no restriction here.

[0057] The virtual vehicle can be understood as a dynamic model vehicle corresponding to the test vehicle in the virtual scene, which is used to simulate the performance of the test vehicle controller in a complex and changing traffic environment.

[0058] Dangerous behaviors can be understood as actions performed by virtual obstacles at the target moment that may pose a threat to vehicle safety, such as sudden acceleration, deceleration, lane changes, pedestrians crossing the road, etc., which are not restricted here.

[0059] Injecting virtual environment information into multiple vehicle sensors and injecting fault signals into multiple sensors at target moments can be understood as using virtual simulation technology to send signals representing the virtual scene to the vehicle's sensors (such as cameras, radars, and lidars). These signals contain the status of all relevant objects (such as virtual obstacles and virtual vehicles) in the virtual scene and environmental conditions (such as weather and lighting). At a specific moment in the virtual scene (the target moment), simulated signals indicating sensor failure are sent to some or all sensors. Fault signals can simulate various sensor issues, such as blurred camera images, radar detection failures, and missing lidar point clouds, to test the autonomous driving system's responsiveness, decision-making capabilities, and system robustness when encountering sensor failures.

[0060] In an embodiment of the present invention, a comprehensive vehicle controller testing environment is constructed by injecting virtual scene information, including virtual obstacles and virtual vehicles, into multiple vehicle sensors in real time. At the target moment when the virtual obstacle performs a dangerous action, simulated fault signals are further injected into the sensors to evaluate the vehicle controller's ability to handle emergency events in the event of sensor failure.

[0061] Step S14, controlling the vehicle controller to generate an execution decision based on the perception information of the multiple sensors after the fault signal is injected and the first posture information, wherein the perception information is data collected from the virtual scene by the multiple sensors after the fault signal is injected;

[0062] In an embodiment of the present invention, the vehicle controller can be understood as the core component of the autonomous driving system, which is responsible for receiving input data from multiple sensors, combining the current posture information of the vehicle, performing data processing, obstacle identification, path planning and decision making through preset algorithms, and finally outputting control instructions to the vehicle's actuators (such as steering system, braking system, power system, etc.) to realize the autonomous driving function.

[0063] Perception information can be understood as data collected from the virtual scene by the vehicle's multiple sensors after receiving virtual environment information and fault signals. For example, perception information can include image data (camera), distance and speed information (radar), and point cloud data (lidar) after fault simulation. This information includes information such as the location, shape, and motion state of obstacles, vehicles, and other environmental features in the virtual scene, without limitation.

[0064] An execution decision can be understood as a specific instruction generated by the vehicle controller based on current perception and vehicle posture information, directing the vehicle's actuators to respond. This can include adjusting vehicle speed, changing direction, or performing emergency braking. The goal is to enable the vehicle to safely navigate various scenarios within the virtual scene, including normal driving, obstacle avoidance, and handling challenges posed by sensor failures.

[0065] Controlling the vehicle controller to generate execution decisions based on the perception information of multiple sensors and the first posture information after the fault signal is injected can be understood as an assessment of the system's ability to make adaptive decisions to ensure safe driving based on limited and potentially damaged environmental perception data combined with the vehicle's real-time posture information when simulating sensor failure in the test of the autonomous driving system.

[0066] In an embodiment of the present invention, based on the received environmental perception data after a simulated sensor failure and the vehicle posture information, safe execution decisions are dynamically generated to ensure that even in the event of a sensor failure, effective responses such as emergency braking or path correction can be made to avoid a collision.

[0067] Step S16 , determining whether the virtual vehicle safely passes through the virtual obstacle based on the execution decision, so as to test the vehicle controller.

[0068] In an embodiment of the present invention, determining whether a virtual vehicle has safely passed through a virtual obstacle based on an execution decision to test the vehicle controller can be understood as the vehicle controller simulating the vehicle in a virtual scene and performing corresponding actions based on the generated execution decision (such as emergency braking, obstacle avoidance steering, etc.), and judging whether the virtual vehicle has safely passed through the virtual obstacle based on the test results.

[0069] In an embodiment of the present invention, the functionality and safety of the vehicle controller are tested by implementing the execution decision generated by the vehicle controller based on fault perception information in a virtual environment and evaluating the obstacle avoidance effect of the virtual vehicle when encountering obstacles.

[0070] In an embodiment of the present invention, by obtaining the first posture information of a vehicle, wherein the vehicle is a vehicle traveling on a real test road; injecting virtual environment information into multiple sensors of the vehicle, and injecting fault signals into the multiple sensors at a target time, wherein the virtual environment information is environmental information of a virtual scene, the virtual scene is used to simulate a vehicle controller test scene, the virtual scene includes a virtual obstacle and a virtual vehicle, the virtual vehicle is used to simulate the driving state of the vehicle in the virtual scene, and the target time is the time when the virtual obstacle performs a dangerous behavior; controlling the vehicle controller to generate an execution decision based on the perception information and the first posture information of the multiple sensors after the fault signal is injected, wherein the perception information is data collected by the multiple sensors from the virtual scene after the fault signal is injected; determining whether the virtual vehicle safely passes through the virtual obstacle based on the execution decision to test the vehicle controller, thereby achieving the purpose of comprehensively testing the fault response capability of the vehicle controller under the collaborative work of multiple sensors, thereby achieving the technical effect of improving the safety and reliability of the autonomous driving vehicle in complex fault scenarios, and thus solving the technical problem of insufficient vehicle stability when only a single sensor is tested in the related art.

[0071] Optionally, in step S12, before injecting fault signals into the multiple sensors at the target time, the vehicle controller testing method further includes the following execution steps:

[0072] Step S121, dynamically adjusting the virtual posture of the virtual vehicle based on the first posture information to obtain first virtual vehicle posture information of the virtual vehicle;

[0073] Step S122: controlling the virtual vehicle to travel in the virtual scene based on the virtual environment information and the first virtual vehicle posture information.

[0074] In an embodiment of the present invention, the first virtual vehicle posture information can be understood as the position, direction, speed, acceleration and posture information of the virtual vehicle that is dynamically updated and reflected in the virtual simulation scene by combining the virtual environment information and the first posture information of the real vehicle (that is, the precise position, direction and motion state of the real vehicle).

[0075] Dynamically adjusting the virtual vehicle's virtual pose based on the first pose information can be understood as using the precise position, orientation, and motion state provided by the real vehicle (i.e., the first pose information) to calibrate the virtual vehicle's position and pose in real time within the virtual environment. This adjustment allows the virtual vehicle to accurately simulate the dynamic behavior of the real vehicle, ensuring pose consistency between the two during testing, thereby improving the authenticity and effectiveness of the test.

[0076] Controlling the virtual vehicle's movement within the virtual scene based on the virtual environment and the first virtual vehicle's positional information can be understood as the simulation system integrating various details of the virtual environment (such as road conditions, weather, and obstacle distribution) with the adjusted first virtual vehicle's positional information to drive the virtual vehicle. This includes simulating the vehicle's acceleration, braking, steering, and other operations, as well as its perception and response to objects in the virtual environment, such as automatically avoiding obstacles and adjusting the vehicle's speed to accommodate slippery roads.

[0077] In an embodiment of the present invention, the virtual vehicle posture is corrected in real time according to the first posture information of the actual vehicle to obtain the first virtual vehicle posture information, and based on the virtual environment and the first virtual posture information, the virtual vehicle is controlled to imitate the driving of the actual vehicle in the scene to achieve precise synchronization, thereby ensuring that the virtual test reflects the actual driving situation, improving the accuracy and reliability of the fault scenario test of the autonomous driving system, and at the same time reducing the cost and risk of actual vehicle testing and accelerating the speed of system iteration and optimization.

[0078] Optionally, in step S14, controlling the vehicle controller to generate an execution decision based on the perception information of the multiple sensors after the fault signal is injected and the first posture information, includes the following execution steps:

[0079] Step S141, controlling the vehicle controller to generate an execution decision based on the sensing information of multiple sensors after the fault signal is injected and the first posture information based on the decision generation algorithm; or,

[0080] Step S142, obtaining the driver's execution operation, wherein the execution operation is the driver's operation of avoiding the virtual obstacle in combination with the display information of the target screen, and the target screen is used to display the virtual scene;

[0081] Step S143: Control the vehicle controller to generate an execution decision based on the decision generation algorithm according to the execution operation, the perception information of multiple sensors after the fault signal is injected, and the first posture information.

[0082] In this embodiment of the present invention, the decision-making algorithm can handle autonomous driving decision-making mechanisms in the event of multiple sensor failures. The decision-making algorithm generates safe and effective execution decisions by analyzing and fusing the perception information of multiple sensors (such as cameras, radars, and lidars) injected with fault signals, as well as the first-level posture information (including position, direction, speed, etc.) provided by the actual vehicle.

[0083] The driver's execution operation can be understood as the driver's action of avoiding virtual obstacles through manual driving (such as turning the steering wheel, stepping on the brake or accelerator) based on the driver's observation of the virtual scene displayed on the target screen in the vehicle-in-the-loop test system.

[0084] The display information of the target screen can be understood as a visual presentation of the virtual scene, including road layout, other vehicles, pedestrians, static obstacles, weather conditions and dynamic events (such as sudden obstacles), etc., which are not restricted here.

[0085] The vehicle controller generates an execution decision based on the perception information and the first posture information of multiple sensors after the fault signal is injected based on the decision generation algorithm. It can be understood that in the automatic driving mode, the vehicle controller uses the decision generation algorithm to analyze the perception information provided by multiple sensors (such as cameras, radars, lidars, etc.) after the interference of the fault signal, and combines the first posture information (including vehicle position, direction and motion status) obtained by the real vehicle to generate execution decisions such as lane changing, deceleration or emergency braking to deal with obstacles or abnormal situations in the virtual scene.

[0086] The driver's execution operation is obtained, wherein the execution operation is the driver's operation of avoiding the virtual obstacle in combination with the display information of the target screen. It can be understood that the driver observes the virtual environment on the target screen, identifies the virtual obstacle, and then manually operates the steering wheel, brake or accelerator, etc. to make an avoidance action.

[0087] The vehicle controller generates an execution decision based on the decision generation algorithm according to the execution operation, the perception information of multiple sensors after the fault signal is injected, and the first posture information. It can be understood that during the test, the vehicle controller performs a comprehensive analysis based on the decision generation algorithm, combined with the driver's operating instructions, the sensor perception information affected by the fault signal, and the first posture information of the actual vehicle to generate the final execution decision.

[0088] In an embodiment of the present invention, the vehicle controller can not only analyze the sensor perception information and actual vehicle posture after fault injection based on the decision-making algorithm to autonomously generate response decisions, but can also obtain the driver's manual operations to avoid obstacles based on the virtual scene displayed on the target screen. By combining the driver's operations with the sensor fault information, the decision generation can be further optimized to ensure the comprehensiveness of the decision and the effectiveness of human-machine collaboration.

[0089] Optionally, in step S16, determining whether the virtual vehicle safely passes through the virtual obstacle based on the execution decision includes the following execution steps:

[0090] Step S161, controlling the vehicle to execute a preset operation based on the execution decision to obtain second posture information of the vehicle, wherein the preset operation is used to avoid a virtual obstacle;

[0091] Step S162, dynamically adjusting the virtual posture of the virtual vehicle based on the second posture information to obtain second virtual vehicle posture information of the virtual vehicle;

[0092] Step S163, determining virtual obstacle posture information of the virtual obstacle based on the second posture information and the perception information;

[0093] Step S164: Based on the second virtual vehicle posture information and the virtual obstacle posture information, determine whether the virtual vehicle safely passes through the virtual obstacle to obtain a determination result.

[0094] In this embodiment of the present invention, a preset action can be understood as a series of actions executed by the vehicle to avoid virtual obstacles based on an execution decision generated by the vehicle controller. Preset actions may include, but are not limited to, deceleration, lane changes, emergency braking, or other driving maneuvers intended to avoid collisions. Preset actions are pre-defined responses to ensure safety in specific fault scenarios.

[0095] The second posture information can be understood as the vehicle's new position, direction, and speed status information obtained in real time by the navigation and positioning system after the vehicle performs the avoidance action.

[0096] The second virtual vehicle posture information can be understood as the second posture information updated in real time. The virtual vehicle in the virtual environment will also adjust its position, direction and speed accordingly to keep it consistent with the current position state of the real vehicle.

[0097] The virtual obstacle pose information can be understood as the position, direction, and status information of the obstacle relative to the vehicle in the virtual scene.

[0098] The determination result can be understood as a test result of determining whether the virtual vehicle successfully and safely avoids the virtual obstacle in the fault scenario based on the comparison between the second virtual vehicle posture information and the virtual obstacle posture information.

[0099] Based on the execution decision, the vehicle is controlled to perform preset operations, and the second posture information of the vehicle is obtained. It can be understood that based on the execution decision, the vehicle is controlled to perform preset operations such as emergency braking, steering avoidance, etc., and the actual position, direction and speed of the vehicle in the road environment are obtained through the navigation and positioning system.

[0100] Dynamically adjusting the virtual posture of the virtual vehicle based on the second posture information to obtain the second virtual vehicle posture information of the virtual vehicle can be understood as follows: in the simulation software, according to the second posture information adjusted by the real vehicle, the posture of the virtual vehicle is also synchronously updated in real time, including the position, direction and speed of the virtual vehicle, to ensure that the virtual test scene is consistent with the real vehicle state.

[0101] Determining the virtual obstacle pose information of the virtual obstacle based on the second pose information and perception information can be understood as the system calculating the position and motion state of the virtual obstacle in the virtual scene based on the latest second pose information of the real vehicle and the perception information of the sensor after fault injection.

[0102] Based on the second virtual vehicle posture information and the virtual obstacle posture information, it is determined whether the virtual vehicle has safely passed the virtual obstacle. The determination result can be understood as comparing the second virtual posture information of the virtual vehicle with the updated virtual obstacle posture information to determine whether the virtual vehicle has successfully avoided collision with the virtual obstacle after performing the avoidance operation, thereby ensuring safe passage.

[0103] In an embodiment of the present invention, a decision is executed to drive the vehicle to perform obstacle avoidance operations, and dynamically updated second posture information is obtained. Subsequently, the position and posture of the vehicle in the virtual environment are adjusted in real time based on this information to ensure synchronization with the actual vehicle state. Then, the posture of the virtual obstacle is estimated by combining the vehicle's second posture information and sensor perception results. Finally, the posture relationship between the virtual vehicle and the obstacle is compared and analyzed to determine whether the obstacle avoidance is successful, forming a test feedback closed loop.

[0104] Optionally, the vehicle controller testing method further includes the following execution steps:

[0105] In response to the determination result indicating that the virtual vehicle safely passed the virtual obstacle, determining that the vehicle controller test passed;

[0106] In response to determining that the virtual vehicle did not safely pass the virtual obstacle, it is determined that the vehicle controller test has failed, and the decision-making algorithm is updated.

[0107] In an embodiment of the present invention, in response to the determination result indicating that the virtual vehicle safely passes through the virtual obstacle, determining that the vehicle controller test has passed can be understood as, when the determination result shows that the virtual vehicle can successfully and safely bypass or pass through the virtual obstacle after performing the obstacle avoidance operation without any collision or dangerous situation, the system determines that the vehicle controller performs well under the tested fault scenario, the decision-making algorithm can effectively deal with the sensor failure, and the overall test result is passed, indicating that the autonomous driving system has stable and reliable decision-making and control capabilities under this type of fault conditions.

[0108] In response to a determination indicating that the virtual vehicle failed to safely pass the virtual obstacle, the vehicle controller test is determined to have failed, and the decision-making algorithm is updated. This means that if the determination indicates that the virtual vehicle failed to safely circumvent or pass the virtual obstacle during an avoidance attempt, collided, or was on the verge of danger, the test result is considered a failure. This indicates that the vehicle controller and its decision-making algorithm have limitations or flaws in specific failure scenarios, requiring algorithm optimization or adjustment to enhance the system's emergency response capabilities and safety in the face of similar sensor failures, ensuring better handling of various complex driving environments and mitigating potential risks in future tests.

[0109] In this embodiment of the present invention, the test status of the vehicle controller is automatically determined based on the dynamic interaction between the virtual vehicle and the obstacle. If the virtual vehicle safely avoids the obstacle, the controller test is confirmed to have passed. Otherwise, the test fails, triggering an optimization update of the decision-making algorithm. This ensures the continuous improvement of the autonomous driving system, especially in complex scenarios such as sensor failures. This not only verifies the current system's capabilities but also provides clear improvement directions for improving system robustness and safety.

[0110] Optionally, the vehicle controller testing method further includes the following execution steps:

[0111] In response to a failure of a first sensor among the plurality of sensors, reducing a weight of the first sensor and increasing a weight of a second sensor, wherein the first sensor is part of the plurality of sensors and the second sensor is a sensor other than the first sensor among the plurality of sensors;

[0112] Determine perception information based on multiple sensors after weight adjustment; or

[0113] In response to a failure of the plurality of sensors, sensing data of the plurality of sensors are fused to obtain fused sensing data;

[0114] The perception information is determined based on the fused perception data.

[0115] In this embodiment of the present invention, the first sensor can be understood as the sensor whose failure is simulated during the test. For example, if the system's performance is being verified when the front-view camera fails, the front-view camera serves as the first sensor. The first sensor can be any one or more sensors in the autonomous driving system, such as lidar, millimeter-wave radar, or visual camera, without limitation.

[0116] The second sensor can be understood as another functioning sensor that the system weights to compensate for missing information when the first sensor fails. For example, in the case of a camera failure, the second sensor might include an unaffected millimeter-wave radar, lidar, or rearview camera, without limitation.

[0117] In response to a failure of the first sensor among multiple sensors, reducing the weight of the first sensor and increasing the weight of the second sensor can be understood as, in a multi-sensor fusion perception system, once a failure of the first sensor (such as a forward-looking camera) is detected, the system immediately adjusts its importance in decision-making, that is, reduces its weight, and at the same time increases the weight of the second sensor (such as a millimeter-wave radar or a lidar) accordingly, to ensure that the decision-making algorithm can rely more on the data of healthy sensors, thereby maintaining the system's perception and decision-making capabilities.

[0118] Determining perception information based on multiple sensors after weight adjustment can be understood as follows: after the above-mentioned weight adjustment, the system recalculates and fuses information from different sensors according to the new weight value of each sensor to generate more accurate and reliable perception information, including object recognition, distance measurement, speed estimation, etc. in the vehicle's surrounding environment, which is the basis for autonomous vehicles to make correct decisions.

[0119] In response to the failure of multiple sensors, the perception data of multiple sensors are fused to obtain fused perception data. It can be understood that in extreme cases, if the system detects that multiple sensors have failed at the same time, it will not simply exclude the faulty data, but will adopt a data fusion strategy. Even if the data quality is damaged, it will try to extract the most valuable information fragments from all available sensor inputs, and through complex algorithm processing (such as Kalman filtering, particle filtering, neural networks, etc.), maximize the restoration and maintenance of the system's perception capabilities, thereby obtaining fused perception data.

[0120] Determining perception information based on fused perception data can be understood as fusing residual data and redundant information based on a data fusion strategy to obtain fused perception data, and judging the driving environment in a timely and effective manner based on the fused perception data to ensure that the vehicle can still maintain a high level of perception and decision-making accuracy in a complex and changing driving environment.

[0121] In the embodiments of the present invention, through intelligent weight adjustment and data fusion, the present invention improves the robustness and decision-making accuracy of the autonomous driving system in the event of sensor failure, effectively avoids the perception blind spots caused by single or multiple sensor failures, ensures driving safety, and promotes the maturity and application of autonomous driving technology.

[0122] Optionally, in step S10, obtaining the first position information of the vehicle includes the following steps:

[0123] The vehicle's position information and posture information are determined based on the vehicle's navigation system and positioning system to obtain the first posture information.

[0124] In the embodiments of the present invention, the navigation system may be understood to refer to an onboard navigation device, which may include a global satellite navigation system (such as GPS, Beidou, Galileo, etc.) receiver for providing the vehicle's geographic coordinates to assist in vehicle positioning and route planning. Furthermore, the navigation system may also include electronic maps and path planning algorithms to help the vehicle understand and predict its driving path, without limitation herein.

[0125] A positioning system can be understood as a system used to continuously track the vehicle's position, speed, and geographic location, including but not limited to an Inertial Navigation System (INS), such as a gyroscope and accelerometer, which can provide near-real-time position updates even in an environment without a Global Positioning System (GPS) signal.

[0126] Position information can be understood as the latitude and longitude coordinates of the vehicle's current location. It is one of the basic inputs for the autonomous driving system to perform path planning, obstacle avoidance, and navigation, enabling the vehicle to accurately understand its precise position in the global coordinate system.

[0127] Attitude information includes the vehicle's attitude angles (pitch, roll, and yaw), as well as dynamic information such as speed and acceleration. Attitude information is crucial for autonomous vehicles, as it involves directional control and stability management, ensuring smooth and safe driving.

[0128] The vehicle's position information and posture information are determined based on the vehicle's navigation system and positioning system. Obtaining the first posture information can be understood as using the vehicle's navigation system (such as GPS) and positioning system (such as INS inertial navigation system) to comprehensively obtain the vehicle's precise geographic location (longitude, latitude, altitude) and dynamic posture (pitch angle, roll angle, yaw angle, as well as dynamic parameters such as speed and acceleration).

[0129] In the embodiment of the present invention, by obtaining the first position information, real-time positioning and direction perception are provided for the autonomous driving vehicle, which is the basis for achieving precise control, path planning and safe obstacle avoidance.

[0130] Figure 2 is a fault injection test scenario diagram according to one embodiment of the present invention, such as Figure 2As shown, the vehicle control system's perception, decision-making, and control capabilities are verified by simulating a complex dynamic scenario, such as a pedestrian suddenly appearing in front of the vehicle, the vehicle's camera being obscured by strong light, and the vehicle's radar signal being lost. First, a virtual scene is constructed in the simulation software. A dynamic target (a pedestrian suddenly appearing between two vehicles) is set, along with light intensity and dynamic changes, to generate a camera halo effect in real time. A virtual vehicle model is then generated based on the vehicle's dimensions and sensor layout, including the fields of view of the forward-facing camera, millimeter-wave radar, and lidar. An injection module is used to block the millimeter-wave radar target signal and simultaneously overlay lidar noise point cloud data to simulate a multi-sensor coordinated failure. Next, environmental data from the simulated scene (such as lighting and pedestrian dynamic trajectory) is transmitted to the forward-facing camera and radar modules via an injection module, generating sensor fault signals. The vehicle's inertial navigation system provides vehicle pose data, which is synchronized to the simulation system in real time to update the vehicle's virtual position in the virtual scene. The fault signal and vehicle position data are then transmitted to the system controller, which receives fused data from the camera, millimeter-wave radar, and lidar. The decision-making algorithm prioritizes target detection results from lidar and cameras based on a multi-sensor weighting mechanism, combining them with the vehicle's position and posture information from inertial navigation and GPS to determine the likely location of obstacles. The vehicle controller outputs an emergency braking signal to the vehicle's actuators, controlling the vehicle to perform the emergency braking maneuver. The inertial navigation system records dynamic parameters (speed, acceleration, and posture changes) during vehicle deceleration. The simulation system simultaneously updates the vehicle's position within the scene, completing the closed-loop scenario.

[0131] Figure 3 FIG. 1 is a structural diagram of a vehicle-in-the-loop system according to one embodiment of the present invention. Figure 3 As shown, the simulation software generates a virtual scene and transmits it to the fault injection module in real time. The fault injection module injects simulated fault signals into each sensor module, and the sensors transmit the sensing data to the vehicle controller. The vehicle controller integrates the multi-sensor data and performs planning and decision-making based on an algorithm. The decision results are transmitted to the vehicle actuators and the vehicle dynamics are fed back to the simulation software. The simulation software adjusts the scene state based on this feedback to maintain synchronization between the simulation and the actual vehicle. The specific implementation steps are as follows: The simulation software generates a virtual scene, including the road environment, dynamic obstacles, and weather conditions. The scene data is output as a video stream and sensor data. The simulation system uses the injection module to superimpose fault signals in the scene in real time, thereby affecting the sensor sensing results and achieving fault injection. The real-time system receives scene information from the simulation software through the host computer and coordinates the synchronous injection of fault signals to ensure consistent timestamps for the multi-sensor data. The injection board converts the fault signals in the simulation system into simulated hardware signals and injects them into the sensor modules.

[0132] The inertial navigation system collects vehicle acceleration and angular velocity, and the GPS records the vehicle's position. The two are combined to generate vehicle pose information. This pose information is transmitted to the simulation software, which dynamically adjusts the position of the virtual vehicle in the scene and is fed back to the system controller to optimize the sensor fault handling algorithm. The driver controls the vehicle's movement using the steering wheel, accelerator, and brakes. This data is transmitted to the system controller, which also affects the response of the vehicle's actuators. The vehicle actuators perform corresponding operations based on the decision instructions of the system controller (automatic braking, steering adjustment, etc.), updating the vehicle's pose in real time. The planning decisions and control signals output by the system controller pass through the vehicle actuators, affecting the vehicle's dynamic performance. The feedback vehicle status (such as speed and acceleration) is transmitted to the simulation system, dynamically adjusting the virtual vehicle's pose in the virtual scene, forming a complete closed loop.

[0133] The present invention breaks through the limitations of traditional single sensor testing, and can simulate the fault states of multiple sensors at the same time in complex scenarios, verifying the decision-making ability of the multi-sensor fusion algorithm under fault conditions. Through vehicle-in-the-loop simulation testing, the system can evaluate the dynamic adjustment ability of the algorithm when the weights of different sensor data change (such as radar failure causing the priority of camera data to increase), ensuring the stability and robustness of the system under fault conditions. The present invention can also build a variety of complex dynamic scenarios in a simulation environment, and flexibly set the fault parameters of sensors such as cameras, radars, and lidars to generate a variety of collaborative fault scenarios, such as a scenario where the camera image is blurred and the radar signal is lost, thereby verifying the comprehensive performance of the system under extreme conditions. By covering low-probability, high-risk scenarios, it ensures that the autonomous driving system still has sufficient robustness under marginal conditions to avoid potential safety hazards.

[0134] In addition, the present invention utilizes vehicle-in-the-loop simulation technology, using virtual scenes to replace real environments, requiring only a small number of real vehicles to participate, significantly reducing site, time, and personnel costs. Through virtual simulation, the present invention can quickly switch and reproduce different test scenarios, support multiple iterative tests, and greatly improve test efficiency. By utilizing real vehicle actuators to participate in the simulation, through closed-loop feedback of perception, planning, decision-making, and execution, test data that is closer to the actual use scenario is generated, thereby improving the authenticity of the test results. At the same time, the simulation system can update the virtual scene in real time according to the vehicle's operating status, ensuring that the vehicle's posture is synchronized with the simulation scene, thereby more realistically reflecting the dynamic impact of multi-sensor failures on the system.

[0135] Through the description of the above embodiments, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus the necessary general hardware platform, and of course it can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for enabling a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods of various embodiments of the present invention.

[0136] This embodiment also provides a vehicle controller testing device for implementing the aforementioned embodiments and preferred implementations. Details already described will not be repeated. As used below, the term "module" may refer to a combination of software and / or hardware that implements a predetermined function. Although the devices described in the following embodiments are preferably implemented using software, implementation using hardware, or a combination of software and hardware, is also possible and contemplated.

[0137] Figure 4 FIG. 1 is a structural block diagram of a vehicle controller testing device according to one embodiment of the present invention. Figure 4 As shown, a vehicle controller test device 400 is taken as an example, and the device includes: an acquisition module 401, the acquisition module is used to obtain the first posture information of the vehicle, wherein the vehicle is a vehicle traveling on a real test road; an injection module 402, the injection module is used to inject virtual environment information into multiple sensors of the vehicle, and inject fault signals into multiple sensors at a target time, wherein the virtual environment information is the environmental information of a virtual scene, the virtual scene is used to simulate the vehicle controller test scene, the virtual scene includes a virtual obstacle and a virtual vehicle, the virtual vehicle is used to simulate the driving state of the vehicle in the virtual scene, and the target time is the time when the virtual obstacle performs a dangerous behavior; a control module 403, the control module is used to control the vehicle controller to generate an execution decision based on the perception information and the first posture information of the multiple sensors after the fault signal is injected, wherein the perception information is the data collected by the multiple sensors from the virtual scene after the fault signal is injected; a test module 404, the test module is used to determine whether the virtual vehicle safely passes through the virtual obstacle based on the execution decision, so as to test the vehicle controller.

[0138] Optionally, before injecting fault signals into the multiple sensors at the target time, the vehicle controller testing device also includes an adjustment module for dynamically adjusting the virtual posture of the virtual vehicle based on the first posture information to obtain the first virtual vehicle posture information of the virtual vehicle; and controlling the virtual vehicle to travel in the virtual scene based on the virtual environment information and the first virtual vehicle posture information.

[0139] Optionally, the control module 403 is also used to control the vehicle controller to generate an execution decision based on a decision generation algorithm according to the perception information of multiple sensors after the fault signal is injected and the first posture information; or, obtain the driver's execution operation, wherein the execution operation is the driver's operation of avoiding the virtual obstacle in combination with the display information of the target screen, and the target screen is used to display the virtual scene; control the vehicle controller to generate an execution decision based on the decision generation algorithm according to the execution operation, the perception information of multiple sensors after the fault signal is injected and the first posture information.

[0140] Optionally, the test module 404 is also used to control the vehicle to perform a preset operation based on the execution decision to obtain second posture information of the vehicle, wherein the preset operation is used to avoid a virtual obstacle; dynamically adjust the virtual posture of the virtual vehicle based on the second posture information to obtain second virtual vehicle posture information of the virtual vehicle; determine the virtual obstacle posture information of the virtual obstacle based on the second posture information and the perception information; determine whether the virtual vehicle has safely passed through the virtual obstacle based on the second virtual vehicle posture information and the virtual obstacle posture information to obtain a determination result.

[0141] Optionally, the vehicle controller testing device also includes a first determination module for determining that the vehicle controller test has passed in response to a determination result indicating that the virtual vehicle has safely passed the virtual obstacle; and for determining that the vehicle controller test has failed in response to a determination result indicating that the virtual vehicle has not safely passed the virtual obstacle, and updating the decision generation algorithm.

[0142] Optionally, the vehicle controller testing device also includes a second determination module for reducing the weight of the first sensor and increasing the weight of the second sensor in response to a failure of the first sensor among multiple sensors, wherein the first sensor is part of the multiple sensors and the second sensor is a sensor other than the first sensor among the multiple sensors; determining perception information based on the multiple sensors after weight adjustment; or in response to a failure of multiple sensors, fusing the perception data of the multiple sensors to obtain fused perception data; and determining perception information based on the fused perception data.

[0143] Optionally, the acquisition module 401 is further configured to determine the position information and posture information of the vehicle based on the navigation system and positioning system of the vehicle to obtain the first posture information.

[0144] It should be noted that the above modules can be implemented through software or hardware. For the latter, it can be implemented in the following ways, but not limited to: the above modules are all located in the same processor; or the above modules are located in different processors in any combination.

[0145] An embodiment of the present invention further provides a vehicle, which is used to execute the steps in any of the above method embodiments.

[0146] An embodiment of the present invention further provides a computer-readable storage medium, in which a computer program is stored. The computer program is configured to execute the steps of any of the above method embodiments when running on a computer or a processor.

[0147] Optionally, in this embodiment, the computer-readable storage medium may be configured to store a computer program for performing the following steps:

[0148] Step S10, obtaining first position information of the vehicle, wherein the vehicle is a vehicle traveling on a real test road;

[0149] Step S12: injecting virtual environment information into multiple sensors of the vehicle and injecting fault signals into the multiple sensors at a target time, wherein the virtual environment information is environmental information of a virtual scene, the virtual scene is used to simulate a vehicle controller test scene, the virtual scene includes a virtual obstacle and a virtual vehicle, the virtual vehicle is used to simulate the driving state of the vehicle in the virtual scene, and the target time is the time when the virtual obstacle performs a dangerous behavior;

[0150] Step S14, controlling the vehicle controller to generate an execution decision based on the perception information of the multiple sensors after the fault signal is injected and the first posture information, wherein the perception information is data collected from the virtual scene by the multiple sensors after the fault signal is injected;

[0151] Step S16 , determining whether the virtual vehicle safely passes through the virtual obstacle based on the execution decision, so as to test the vehicle controller.

[0152] Optionally, in this embodiment, the above-mentioned computer-readable storage medium may include but is not limited to: a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk, and other media that can store computer programs.

[0153] An embodiment of the present invention further provides an electronic device, comprising a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0154] Optionally, in this embodiment, the processor in the electronic device may be configured to run a computer program to perform the following steps:

[0155] Step S10, obtaining first position information of the vehicle, wherein the vehicle is a vehicle traveling on a real test road;

[0156] Step S12: injecting virtual environment information into multiple sensors of the vehicle and injecting fault signals into the multiple sensors at a target time, wherein the virtual environment information is environmental information of a virtual scene, the virtual scene is used to simulate a vehicle controller test scene, the virtual scene includes a virtual obstacle and a virtual vehicle, the virtual vehicle is used to simulate the driving state of the vehicle in the virtual scene, and the target time is the time when the virtual obstacle performs a dangerous behavior;

[0157] Step S14, controlling the vehicle controller to generate an execution decision based on the perception information of the multiple sensors after the fault signal is injected and the first posture information, wherein the perception information is data collected from the virtual scene by the multiple sensors after the fault signal is injected;

[0158] Step S16 , determining whether the virtual vehicle safely passes through the virtual obstacle based on the execution decision, so as to test the vehicle controller.

[0159] An embodiment of the present invention further provides a computer program product, including a computer program, which implements the steps of any of the above method embodiments when executed by a processor.

[0160] Optionally, in this embodiment, the computer program in the above computer program product may be configured to perform the following steps when executed by a processor:

[0161] Step S10, obtaining first position information of the vehicle, wherein the vehicle is a vehicle traveling on a real test road;

[0162] Step S12: injecting virtual environment information into multiple sensors of the vehicle and injecting fault signals into the multiple sensors at a target time, wherein the virtual environment information is environmental information of a virtual scene, the virtual scene is used to simulate a vehicle controller test scene, the virtual scene includes a virtual obstacle and a virtual vehicle, the virtual vehicle is used to simulate the driving state of the vehicle in the virtual scene, and the target time is the time when the virtual obstacle performs a dangerous behavior;

[0163] Step S14, controlling the vehicle controller to generate an execution decision based on the perception information of the multiple sensors after the fault signal is injected and the first posture information, wherein the perception information is data collected from the virtual scene by the multiple sensors after the fault signal is injected;

[0164] Step S16 , determining whether the virtual vehicle safely passes through the virtual obstacle based on the execution decision, so as to test the vehicle controller.

[0165] Optionally, specific examples in this embodiment may refer to the examples described in the above embodiments and optional implementation modes, and this embodiment will not be described in detail here.

[0166] The serial numbers of the above embodiments of the present invention are for description only and do not represent the advantages or disadvantages of the embodiments.

[0167] In the above embodiments of the present invention, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0168] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only exemplary. For example, the division of units can be a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0169] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0170] In addition, the functional units in the various embodiments of the present invention may be integrated into a single processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0171] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM), random access memory (RAM), mobile hard disk, magnetic disk or optical disk, etc., various media that can store program code.

[0172] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as within the scope of protection of the present invention.

Claims

1. A vehicle controller testing method, characterized in that: include: Acquiring first position information of a vehicle, wherein the vehicle is a vehicle traveling on a real test road; injecting virtual environment information into multiple sensors of the vehicle and injecting fault signals into the multiple sensors at a target time, wherein the virtual environment information is environmental information of a virtual scene, the virtual scene is used to simulate a vehicle controller test scene, the virtual scene includes a virtual obstacle and a virtual vehicle, the virtual vehicle is used to simulate the driving state of the vehicle in the virtual scene, and the target time is the time when the virtual obstacle performs a dangerous behavior; Controlling the vehicle controller to generate an execution decision based on the perception information of the multiple sensors and the first posture information, wherein the perception information is data collected by the multiple sensors from the virtual scene after the fault signal is injected; The vehicle controller is tested by determining whether the virtual vehicle safely passes the virtual obstacle based on the execution decision.

2. The method according to claim 1, characterized in that Before injecting fault signals into the plurality of sensors at a target time, the method further includes: Dynamically adjusting the virtual posture of the virtual vehicle based on the first posture information to obtain first virtual vehicle posture information of the virtual vehicle; The virtual vehicle is controlled to travel in the virtual scene based on the virtual environment information and the first virtual vehicle posture information.

3. The method according to claim 1, characterized in that The controlling the vehicle controller to generate an execution decision based on the perception information of the multiple sensors after the fault signal is injected and the first posture information, including: Based on a decision generation algorithm, controlling the vehicle controller to generate the execution decision according to the perception information of the multiple sensors and the first posture information; or, Acquiring the driver's execution operation information, wherein the execution operation information is the driver's operation information for avoiding the virtual obstacle in combination with display information on a target screen, the target screen being used to display the virtual scene; Based on the decision generation algorithm, the vehicle controller is controlled to generate the execution decision according to the execution operation information, the perception information of the multiple sensors and the first posture information.

4. The method according to claim 1, wherein The determining, based on the execution decision, whether the virtual vehicle safely passes through the virtual obstacle comprises: controlling the vehicle to execute a preset operation based on the execution decision to obtain second posture information of the vehicle, wherein the preset operation is used to avoid the virtual obstacle; Dynamically adjusting the virtual posture of the virtual vehicle based on the second posture information to obtain second virtual vehicle posture information of the virtual vehicle; determining virtual obstacle pose information of the virtual obstacle based on the second pose information and the perception information; Based on the second virtual vehicle position information and the virtual obstacle position information, it is determined whether the virtual vehicle safely passes through the virtual obstacle.

5. The method according to claim 4, characterized in that The method further comprises: In response to the virtual vehicle safely passing the virtual obstacle, determining that the vehicle controller test passes; In response to the virtual vehicle failing to safely pass the virtual obstacle, it is determined that the vehicle controller test has failed, and a decision-making algorithm is updated.

6. The method according to claim 1, characterized in that The method further comprises: In response to a failure of a first sensor among the plurality of sensors, reducing the weight of the first sensor and increasing the weight of a second sensor, wherein the first sensor is part of the plurality of sensors and the second sensor is a sensor among the plurality of sensors other than the first sensor; Determine the perception information based on the plurality of sensors after weight adjustment; or In response to a failure of all of the multiple sensors, fusing sensing data from the multiple sensors to obtain fused sensing data; The perception information is determined based on the fused perception data.

7. The method according to claim 1, characterized in that The obtaining of the first position information of the vehicle includes: The position information and posture information of the vehicle are determined based on the navigation system and positioning system of the vehicle to obtain the first posture information.

8. A vehicle controller testing device, characterized in that: The device comprises: An acquisition module, configured to acquire first position information of a vehicle traveling on a real test road; an injection module, the injection module being configured to inject virtual environment information into a plurality of sensors of the vehicle and to inject fault signals into the plurality of sensors at a target time, wherein the virtual environment information is environmental information of a virtual scene, the virtual scene being configured to simulate a vehicle controller test scene, the virtual scene comprising a virtual obstacle and a virtual vehicle, the virtual vehicle being configured to simulate a driving state of the vehicle in the virtual scene, and the target time being the time at which the virtual obstacle performs a dangerous behavior; a control module, the control module being configured to control the vehicle controller to generate an execution decision based on perception information of the multiple sensors after the fault signal is injected and the first posture information, wherein the perception information is data collected from the virtual scene by the multiple sensors after the fault signal is injected; A testing module is used to determine whether the virtual vehicle safely passes through the virtual obstacle based on the execution decision, so as to test the vehicle controller.

9. A computer-readable storage medium, characterized in that The storage medium stores a computer program, wherein the computer program is configured to execute the vehicle controller testing method according to any one of claims 1 to 7 when running on a computer or a processor.

10. An electronic device comprising a memory and a processor, characterized in that: The memory stores a computer program, and the processor is configured to run the computer program to execute the vehicle controller testing method according to any one of claims 1 to 7.

11. A computer program product, characterized in that The invention comprises a computer program, which implements the vehicle controller testing method according to any one of claims 1 to 7 when being executed by a processor.

Images