Automobile left-right parking adjustable scene automatic test device and test method

Abstract

The invention provides an automatic test device and a test method for an adjustable scene of left and right parking of an automobile, and belongs to the technical field of intelligent driving test. The test device comprises a test upper computer used for outputting a first control instruction and a second control instruction in response to a test instruction of a user; the multi-degree-of-freedom dynamic scene generation system is used for receiving the first control instruction and generating test scenes on the left side and the right side of the tested vehicle based on the first control instruction; the test control terminal is used for receiving the second control instruction and triggering the parking-out or parking-in operation of the tested vehicle based on the second control instruction; the position sensing device is used for collecting position data of a detected vehicle and an obstacle in real time; the test upper computer is also used for generating a test result based on the position data. According to the device, through integrating the test upper computer, the dynamic scene generation system and the position sensing device, automatic scene construction is realized, the test efficiency is improved, and the device has the advantages of realizing standardized test conditions and covering more boundary scenes.

Description

Technical Field

[0001] This invention belongs to the field of intelligent driving testing technology, specifically relating to an automated testing device and method for adjustable left and right parking scenarios of automobiles. Background Technology

[0002] With the deepening development of vehicle electrification and intelligence, models equipped with Automatic Parking Assist (APA) are becoming increasingly popular. Early automatic parking exit functions mainly focused on straight-line exits, but to cope with complex urban parking environments, left and right parking exit functions have gradually become the mainstream. This function requires the vehicle to be able to autonomously choose to exit to the left or right based on instructions, and also to choose to drive away to the left or right based on driver instructions. The realization of left and right parking exit functions places higher demands on the perception, decision-making, and control capabilities of intelligent driving systems.

[0003] In real-world usage scenarios, the left and right parking exit function is prone to decision-making errors in dynamically changing environments, leading to parking failures or even collisions, posing safety risks and potential property damage. Therefore, it is essential to conduct thorough and reliable testing and verification of the left and right parking exit function before vehicle delivery.

[0004] Currently, the industry primarily relies on manual field testing for functional verification. Testers need to find suitable parking spaces in real parking lots or manually arrange obstacles to simulate test scenarios. This method has the following significant drawbacks: insufficient scenario coverage, high randomness, and inability to systematically and accurately cover all the boundary and extreme conditions that need to be verified; manually constructed scenarios are difficult to systematically reproduce boundary and extreme conditions, resulting in poor consistency of test conditions, and inconsistencies in the timing of manual triggering commands by engineers. This lack of standardized operation timing leads to non-standardized test conditions, making it difficult to reproduce and locate the root cause when problems occur; the test process relies on manual labor, resulting in low efficiency and failing to meet the needs of large-scale regression testing with rapid iteration; furthermore, manual testing is difficult to safely and controllably simulate dynamic adversarial scenarios, posing high operational risks and limitations. In summary, the shortcomings of existing testing methods in terms of automation, standardization, and scenario controllability have become key bottlenecks restricting the development and verification of left and right parking functions.

[0005] Therefore, how to achieve automated, standardized, and scenario-controllable testing has become an urgent problem to be solved. Summary of the Invention

[0006] The purpose of this invention is to solve the problems existing in the prior art and provide an automated testing device and method for adjustable left and right parking scenarios of automobiles, which has the advantages of improving testing efficiency, realizing standardized testing conditions, and covering more boundary scenarios.

[0007] This invention is achieved through the following technical solution:

[0008] The first aspect of the invention provides an automated testing device for adjustable parking scenarios of automobiles, comprising: a testing host computer, a multi-degree-of-freedom dynamic scene generation system, a testing control terminal, and a position sensing device, wherein the testing host computer is communicatively connected to the multi-degree-of-freedom dynamic scene generation system, the testing control terminal, and the position sensing device, respectively.

[0009] The host computer for testing is used to respond to the user's test instructions, output a first control instruction to control the multi-degree-of-freedom dynamic scene generation system to generate a corresponding test scene; and output a second control instruction to control the test control terminal to trigger a vehicle parking operation.

[0010] The multi-degree-of-freedom dynamic scene generation system is used to receive the first control command and generate corresponding test scenes on the left and right sides of the vehicle under test based on the first control command.

[0011] The test control terminal is used to receive the second control command and trigger the parking out or parking in operation of the vehicle under test based on the second control command.

[0012] The position sensing device is used to collect position data of the vehicle and obstacles under test in real time and send the collected position data to the test host computer.

[0013] The host computer for testing is also used to receive the location data and generate test results based on the location data.

[0014] Optionally, the multi-degree-of-freedom dynamic scene generation system includes a motion platform, the motion platform comprising:

[0015] A two-dimensional moving mechanism is used to move along the X and Y axes in a horizontal plane to simulate the obstacle;

[0016] A rotating turntable is positioned below the two-dimensional moving mechanism to support the two-dimensional moving mechanism and enable rotational adjustment.

[0017] A motion controller is used to receive the first control command and control the movement of the two-dimensional moving mechanism and the rotating turntable.

[0018] Optionally, the rotation angle of the rotating turntable ranges from 0 to 180 degrees, used to simulate angled parking spaces or adjacent vehicles with incorrect posture.

[0019] Optionally, the test control terminal is a smartphone, which communicates with the test host computer via the ADB protocol; the smartphone is used to receive ADB commands sent by the test host computer to automate the operation of the vehicle control APP.

[0020] Optionally, the position sensing device includes a lidar and / or an image acquisition device, used to acquire the contour information and position information of the vehicle under test and the obstacle through point cloud analysis and / or image processing.

[0021] Optionally, the test host computer is used for:

[0022] Control the deployment of the initial parking space scene by the multi-degree-of-freedom dynamic scene generation system;

[0023] During the parking process of the vehicle under test, the dynamic scene generation system is synchronously controlled to perform corresponding dynamic changes;

[0024] Based on the data returned by the position sensing device, it is determined in real time whether the tested vehicle has collided or failed to park.

[0025] After the test is completed, the vehicle under test is controlled to return to the initial parking space, and a test report is generated.

[0026] A second aspect of the present invention provides an automated testing method for left and right parking of a vehicle, executed by a host computer in the device described in any one of the first aspects above, comprising:

[0027] Receive test commands from users;

[0028] In response to the test command, the multi-degree-of-freedom dynamic scene generation system is controlled to generate corresponding test scenes on the left and right sides of the vehicle under test by outputting a first control command;

[0029] In response to the test command, the test control terminal is controlled by outputting a second control command to trigger the parking out or parking in operation of the vehicle under test;

[0030] The system receives real-time position data of the vehicle and obstacles under test from the position sensing device and generates test results based on the position data.

[0031] Optionally, the test instructions include static scene parameters, berthing direction, and dynamic scene sequence.

[0032] Optionally, corresponding test scenarios are generated on the left and right sides of the vehicle under test, including:

[0033] Based on the static scene parameters, the multi-degree-of-freedom dynamic scene generation system is controlled to deploy the initial parking space scene;

[0034] During the parking process of the vehicle under test, the dynamic scene generation system is synchronously controlled to perform corresponding dynamic changes based on the dynamic scene sequence.

[0035] Optionally, the dynamic scene sequence includes controlling the obstacle to move along a preset trajectory during the parking process of the vehicle under test, so as to simulate the dynamic behavior of the vehicle next to it.

[0036] Compared with the prior art, the beneficial effects of the present invention are as follows: The automated testing device and method for adjustable parking scenarios of automobiles provided in this application realizes automated scenario construction and real-time data feedback by integrating a test host computer, a dynamic scenario generation system and a position sensing device. It solves the technical problems of low efficiency and uncontrollable scenarios in manual testing, and has the advantages of improving testing efficiency, realizing standardized testing conditions, covering more boundary scenarios and improving testing safety. Attached Figure Description

[0037] The above and other objects, features, and advantages of the present invention will become more apparent from the more detailed description of the embodiments of the invention in conjunction with the accompanying drawings. The drawings are provided to further illustrate the embodiments of the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof.

[0038] Figure 1 This is a schematic diagram of the structure of an automated testing device for adjustable left and right parking scenarios of a car, provided in an embodiment of the present invention.

[0039] Figure 2 A top view of a multi-degree-of-freedom dynamic scene generation system provided in an embodiment of the present invention;

[0040] Figure 3 A flowchart illustrating an automated testing method for left and right parking of a vehicle, provided in an embodiment of the present invention;

[0041] Figure 4 The flowchart illustrates the operation of an automated testing device for adjustable left and right parking scenarios for automobiles, as provided in this embodiment of the invention. Detailed Implementation

[0042] To make the objectives, technical solutions, and advantages of the present invention more apparent, exemplary embodiments according to the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are merely a part of the embodiments of the present invention, and not all of the embodiments of the present invention. It should be understood that the present invention is not limited to the exemplary embodiments described herein. Based on the embodiments of the present invention described herein, all other embodiments obtained by those skilled in the art without inventive effort should fall within the protection scope of the present invention.

[0043] In existing technologies, with the rapid development of intelligent driving technology for automobiles, automatic parking functions have gradually expanded from single-direction to bidirectional operation. Traditional testing methods rely on manually setting up obstacles on-site and manually triggering vehicle operation, which suffers from problems such as incomplete scene coverage, inconsistent test conditions, difficulty in simulating dynamically changing scenes, and low testing efficiency. Therefore, this application proposes an automated testing device and method for adjustable left and right parking scenarios for automobiles, which will be described in detail below with reference to the accompanying drawings.

[0044] First, refer to Figure 1 This invention describes an automated testing device for adjustable left and right parking scenarios for automobiles, provided by an embodiment of the present invention.

[0045] like Figure 1 As shown, the automated testing device for adjustable parking scenarios of cars in this embodiment of the invention includes: a test host computer, a multi-degree-of-freedom dynamic scene generation system, a test control terminal, and a position sensing device. The test host computer is communicatively connected to the multi-degree-of-freedom dynamic scene generation system, the test control terminal, and the position sensing device, respectively.

[0046] The host computer for testing is used to respond to the user's test instructions, output a first control instruction to control the multi-degree-of-freedom dynamic scene generation system to generate a corresponding test scene; and output a second control instruction to control the test control terminal to trigger a vehicle parking operation.

[0047] The multi-degree-of-freedom dynamic scene generation system is used to receive the first control command and generate corresponding test scenes on the left and right sides of the vehicle under test based on the first control command.

[0048] The test control terminal is used to receive the second control command and trigger the parking out or parking in operation of the vehicle under test based on the second control command.

[0049] The position sensing device is used to collect position data of the vehicle and obstacles under test in real time and send the collected position data to the test host computer.

[0050] The host computer for testing is also used to receive the location data and generate test results based on the location data.

[0051] The test host computer refers to the central control unit, which can be implemented using an industrial control computer, coordinating the work of various subsystems through communication protocols. The multi-degree-of-freedom dynamic scene generation system refers to a programmable motion device used to precisely deploy obstacle layouts on both sides of the vehicle. The test control terminal refers to the vehicle operation triggering device, which automates the operation of the vehicle control APP through simulated touch commands. The position sensing device refers to spatial positioning equipment, such as a combination of LiDAR and industrial cameras, which obtains the real-time positional relationship between the vehicle and obstacles through three-dimensional coordinate calculation.

[0052] Specifically, when a user initiates a test command, the host computer simultaneously generates scenario construction instructions and vehicle operation instructions. The scenario generation system deploys initial obstacle positions on both sides of the vehicle according to preset parameters, while the test control terminal triggers the vehicle parking function via a communication protocol. During vehicle movement, position sensors collect spatial coordinate data at a fixed frequency and transmit it to the host computer in real time for collision detection and trajectory analysis. The host computer automatically determines the success or failure of the parking operation by comparing the actual trajectory with the expected path and generates a test report including parameters such as timestamps and offsets.

[0053] Compared to existing technologies, traditional manual testing methods require testers to repeatedly adjust obstacle positions in a real-world environment. This solution, however, utilizes a programmable motion device to achieve precise scene reproduction and dynamic adjustment. Existing technologies rely on manual visual judgment of vehicle trajectories; this solution employs high-precision sensing technology to achieve millimeter-level positioning measurements. Existing testing processes suffer from timing errors in vehicle operation triggering; this solution ensures consistent command execution timing through a standardized communication protocol.

[0054] This invention provides an automated testing device for adjustable left and right parking scenarios, integrating a multi-degree-of-freedom dynamic scene generation system, a host computer, a test control mobile phone, and a position sensing device. It achieves automated deployment of left and right parking scenarios under simulated conditions, accurately simulating different parking scenarios on both sides of the vehicle. The automated operation process eliminates random errors inherent in manual testing, ensuring consistency of test conditions. Real-time position monitoring and automatic result judgment mechanisms significantly improve testing efficiency and result reliability. The dynamic scene generation system supports synchronous adjustment of obstacle positions during parking, effectively verifying the vehicle's real-time decision-making capabilities in complex environments.

[0055] In some embodiments of the present invention, the position sensing device includes a lidar and / or an image acquisition device for acquiring contour information and position information of the vehicle under test and the obstacle through point cloud analysis and / or image processing.

[0056] Position sensing devices acquire the contour and position information of the tested vehicle and obstacles in real time and with high accuracy through non-contact measurement, providing a data basis for the automatic judgment of the host computer. For example, the image acquisition device can be a high-performance industrial camera.

[0057] When the position sensing device includes a lidar, its workflow is as follows:

[0058] Data acquisition: The lidar is fixed above or at a key position in the test area. Its emitted laser beam scans the test area at high speed and receives the returned signal, thereby generating dense environmental point cloud data containing three-dimensional coordinate information.

[0059] Point cloud analysis and processing: Testing the dedicated processing software in the host computer to process point cloud data in real time:

[0060] Point cloud clustering and segmentation: First, the point clouds belonging to the tested vehicles and obstacles are separated from the ground point clouds and background noise by algorithms (such as Euclidean clustering and region growing segmentation) to form independent point cloud clusters.

[0061] Contour extraction and pose calculation: For each point cloud cluster, the software accurately obtains the physical contour of the target by calculating its 3D bounding box or fitting its external contour. Furthermore, by analyzing the spatial distribution of the point cloud cluster, the real-time position (usually represented by the coordinates of the center point of the bounding box (X,Y,Z)) and attitude (i.e., yaw angle, pitch angle, and roll angle) of the target can be calculated.

[0062] Collision judgment criteria: By continuously calculating the minimum distance between the vehicle contour point cloud and the obstacle contour point cloud, it is possible to determine in real time whether there is a collision risk or whether a scrape has occurred.

[0063] When the position sensing device includes an image acquisition device, its workflow is as follows:

[0064] Data acquisition: One or more industrial cameras view the entire test area from above, capturing high-resolution digital images. To ensure measurement accuracy, the cameras need to be calibrated beforehand to obtain their intrinsic parameters (such as focal length and distortion coefficient) and extrinsic parameters (position and orientation in the world coordinate system).

[0065] Image Processing and Recognition: Testing the host computer's processing of the acquired images:

[0066] Target detection and recognition: Using deep learning-based target detection algorithms (such as YOLO, SSD, Faster R-CNN, etc.), the detected vehicles and obstacles are quickly and accurately identified in the image, and their pixel regions (bounding boxes) are marked.

[0067] Coordinate transformation and positioning: Combining camera calibration parameters and utilizing perspective projection geometry, the pixel coordinates of the target in the 2D image are transformed into 3D coordinates (X, Y) in a predefined world coordinate system. If a binocular or multi-view stereo vision system is used, the target's depth information (Z coordinate) can be directly obtained through parallax calculation, achieving 3D positioning.

[0068] Contour and attitude analysis: By analyzing the boundary contour features of the target, its attitude angles can be further estimated.

[0069] In a preferred embodiment, the position sensing device includes both a LiDAR and an industrial camera, forming a multi-sensor fusion perception system. This system uses a data fusion algorithm to align and fuse the precise 3D point cloud data provided by the LiDAR with the rich texture and color information provided by the industrial camera. This fusion scheme combines the advantages of LiDAR's accurate ranging and immunity to lighting conditions with the strengths of visual perception in object classification and contour detail resolution, generating more reliable and complete environmental perception results.

[0070] Workflow: The fusion system can use the 3D information provided by LiDAR point clouds as a foundation, and then use visual images to perform semantic segmentation on the point clouds (e.g., to more accurately distinguish different parts of a vehicle), or use visual recognition results to assist the point clouds in performing faster and more accurate target classification. Ultimately, the system outputs a high-precision environment model integrating contours, position, pose, and even object type.

[0071] Whether using a single sensor or a fusion scheme, the position sensing device can achieve real-time, high-precision positioning and monitoring of the core moving objects (vehicles and obstacles) in this testing device, providing indispensable data support for the judgment of automated test results.

[0072] The following combination Figure 2 This document provides a detailed description of the automated testing device for adjustable parking scenarios for automobiles.

[0073] The automated testing device for adjustable parking scenarios of automobiles consists of a host computer, a multi-degree-of-freedom dynamic scene generation system, a test control mobile phone, and a position sensing device. The host computer is communicatively connected to the multi-degree-of-freedom dynamic scene generation system, the test control mobile phone, and the position sensing device.

[0074] The host computer serves as the control center of the entire testing system, integrating test control software. When the software program runs, it can control all hardware devices. The host computer is responsible for processing test logic, sending control commands, receiving and processing sensor data, and ultimately completing data analysis and test report generation.

[0075] The multi-degree-of-freedom dynamic scene generation system is the core actuator for realizing the automatic generation and dynamic changes of physical scenes. In this embodiment, the multi-degree-of-freedom dynamic scene generation system includes an XYR three-degree-of-freedom motion platform, which comprises two two-dimensional moving mechanisms, a rotary table, and a motion controller. The two-dimensional moving mechanisms are used to move along the X and Y axes in the horizontal plane to simulate obstacles; the rotary table is located below the two-dimensional moving mechanisms to support them and achieve rotational adjustment; the motion controller receives first control commands and controls the movement of the two-dimensional moving mechanisms and the rotary table.

[0076] like Figure 2 As shown, specifically, the two-dimensional movement mechanism consists of two XY two-dimensional translation slide rails 21 and 22, which are mounted on a rotating turntable 10 that can rotate around the R-axis and are respectively arranged on both sides of the turntable 10. The area in the middle of the slide rails is the parking area of ​​the vehicle under test.

[0077] Two-dimensional translation rails 21 and 22 can move within three degrees of freedom (X, Y, and R) as obstacles: the rotating turntable 10 can rotate precisely from 0 to 180 degrees around the R axis, and the two-dimensional translation rails 21 and 22 can rotate with the rotating turntable 10, thus simulating angled parking spaces at any angle, or simulating adjacent vehicles that are not in the correct posture or not parked correctly; the two-dimensional translation rails 21 and 22 can simulate obstacles such as vehicles and walls at different positions next to the vehicle by moving forward, backward, left, and right.

[0078] The XYR three-degree-of-freedom motion platform has the ability to move along the Y-axis longitudinally and the X-axis transversely, and is also driven by high-precision servo motors. This platform can not only automatically deploy static obstacles at any position and angle (simulating diagonal parking spaces and non-standard parking spaces), but also dynamically move obstacles according to a preset trajectory during test execution.

[0079] The motion controller receives instructions from the host computer and coordinates the rotation angle of the turntable as well as the motion trajectory, speed, and position of the X and Y axis slide rails.

[0080] In this embodiment, the multi-degree-of-freedom dynamic scene generation system can deploy the initial parking space scene under the control of the test host computer; during the parking process of the vehicle under test, the dynamic scene generation system can execute corresponding dynamic changes.

[0081] In this embodiment, the test control terminal is a smartphone with a remote control APP installed on it, compatible with the vehicle under test. The smartphone connects to the host computer via a USB cable, serving as the execution terminal for automated operations. It receives and executes Android Debug Bridge (ADB) commands sent by the host computer, automating the clicking of left and right parking functions in the smartphone control APP, thereby triggering the relevant parking operations of the vehicle under test. Through automated test triggering based on the test smartphone's ADB, the left and right parking, automatic parking, and other functions of the vehicle under test can be triggered accurately and repeatedly.

[0082] Please continue reading. Figure 2 In this embodiment, position sensing devices 31, 32, 33, and 34 are used to accurately monitor and determine the real-time position and attitude of the tested vehicle and obstacles, and to automatically determine the results. The position sensing devices may include lidar and industrial cameras installed above the test area at both ends of the two-dimensional translation rails 21 and 22, which capture images of the entire test area from a high vantage point and obtain the precise coordinates and contours of the vehicle and obstacles in real time through image processing or point cloud analysis algorithms.

[0083] The host computer receives position data sent by the position sensing device during the test, and determines whether there are collisions, scrapes, or parking failures during the left and right parking process based on the position of the vehicle under test and the multi-degree-of-freedom dynamic scene generation system. After the test is completed, the host computer controls the vehicle under test to return to the initial parking space and generates a test report.

[0084] The automated testing device for adjustable left and right parking scenarios provided in this embodiment can automatically generate and adjust parking spaces by controlling a multi-degree-of-freedom dynamic scene generation system via a host computer. The host computer uses ADB commands to control a mobile app for automatic parking and left / right parking control. During left and right parking, the host computer can also simultaneously control the left and right sliding rails on both sides of the vehicle to simulate the movement of vehicles on either side. The device uses position sensors to automatically judge and monitor the results of the entire parking process. Finally, the host computer generates a test report based on the results of each automatic judgment. This device has the following advantages:

[0085] I. Improve testing efficiency and shorten the R&D cycle

[0086] This automated testing system boasts a simple architecture, low manufacturing cost, and easy maintenance. It automates the entire process from scenario generation, test execution, result evaluation, to report output. Controlled by a host computer program, it can perform 24 / 7 uninterrupted testing of hundreds of different scenario test cases, completely eliminating the time-consuming and labor-intensive manual scenario setup, triggering, and data analysis required in traditional testing. This significantly reduces test waiting and execution times, meeting the large-scale regression testing needs of modern automotive software's rapid iteration, thereby drastically shortening product development and verification cycles.

[0087] II. Achieving full scenario coverage and precise control of testing

[0088] This embodiment utilizes a multi-degree-of-freedom dynamic scene generation system to accurately and quantitatively create complex static parking spaces of arbitrary width, angle, and rear-end misalignment, effectively covering boundary and extreme conditions that are difficult to find in the real world. More importantly, it can accurately simulate high-risk dynamic confrontation scenarios such as "movement of adjacent vehicles during parking," which is impossible to achieve safely and accurately with traditional manual testing. Through the automated triggering of ADB commands, it ensures that the command triggering and dynamic changes of the scene are precisely synchronized and controlled for each test, avoiding the randomness of manual operation.

[0089] III. Ensuring high repeatability and reliability of the test

[0090] This automated testing device ensures strict consistency in scenario parameters, dynamic event timing, and trigger commands for each test through programmed control. When a scenario causing failure is detected, it can be reproduced with 100% accuracy, providing R&D engineers with a stable and reliable basis for locating and resolving problems. This unparalleled repeatability completely solves the industry pain point of "intermittent and difficult-to-reproduce problems" caused by the inability to accurately reproduce scenarios and operations in manual testing.

[0091] IV. Comprehensively improve product safety and quality

[0092] This device can systematically and exhaustively verify the left and right parking functions of vehicles under various complex, dynamic, and extreme scenarios, effectively exposing potential defects in the perception, decision-making, or control levels of autonomous driving systems. By conducting thorough and rigorous automated testing during the R&D phase, safety accidents such as scratches and collisions caused by parking function failure can be avoided, thereby significantly reducing related after-sales claims and brand reputation risks, and ensuring that the products delivered to users have higher safety and reliability.

[0093] Next, refer to Figure 3 This invention describes an automated testing method for left and right parking of a vehicle, which is executed by a host computer in the device described in any of the above embodiments.

[0094] like Figure 3 As shown, the automated testing method for parking vehicles to the left and right includes the following steps S100-S400:

[0095] Step S100: Receive the user's test command;

[0096] Step S200: In response to the test command, the multi-degree-of-freedom dynamic scene generation system is controlled to generate corresponding test scenes on the left and right sides of the vehicle under test by outputting a first control command;

[0097] Step S300: In response to the test command, the test control terminal is controlled to trigger the parking or parking operation of the vehicle under test by outputting a second control command;

[0098] Step S400: Receive the position data of the vehicle under test and obstacles collected in real time by the position sensing device, and generate test results based on the position data.

[0099] Among them, test instructions refer to the test parameters and operation instructions set by the user. Specifically, they can be implemented through input via a graphical interface or by calling preset scripts, and are used to trigger the standardized execution of automated test processes.

[0100] Specifically, after the test process begins, the host computer receives the test commands input by the user and parses the test parameters contained within. The multi-degree-of-freedom dynamic scene generation system constructs parking space environments on both sides of the vehicle based on the test parameters. The test control terminal sends a parking command to the vehicle through a preset communication protocol, triggering the automatic parking function. The position sensing device continuously collects real-time position data of the vehicle and obstacles, and calculates the distance and relative motion trend between them through a coordinate transformation algorithm. Based on a preset collision threshold and path planning model, the host computer analyzes the position data to determine whether deviation from the path, collision risk, or operation timeout occurred during the parking process, and finally generates a quantitative test report containing the test pass rate, trajectory deviation, and abnormal events.

[0101] Through the above technical solutions, this application solves the problem of insufficient coverage of manual testing scenarios, achieving accurate reproduction of parking space layout parameters and obstacle dynamic behavior, covering complex working conditions such as angled parking spaces and moving obstacles. It eliminates fluctuations in test conditions caused by inconsistent operation timing, ensuring strict synchronization between vehicle triggering timing and scenario changes in each test. An automated evaluation mechanism based on quantitative data is established, avoiding the subjectivity of human judgment and improving the reliability and traceability of test results.

[0102] In one embodiment of the present invention, the test instruction includes static scene parameters, berthing direction, and dynamic scene sequence.

[0103] Static scene parameters refer to the initial configuration data used to define fixed elements in the test scene. Specifically, this can be achieved using parking space length, distance between adjacent obstacles, and roadside position coordinates. Standardizing the geometric parameters of the initial scene ensures consistent baseline conditions for each test. Parking exit direction refers to the directional command controlling the test vehicle's departure from the parking space. This can be implemented using left- or right-direction binary encoded signals. By clearly defining the vehicle's movement direction, the system's response logic under different decision paths is verified. Dynamic scene sequences refer to a set of obstacle movement rules arranged chronologically. Specifically, this can be achieved using obstacle movement trajectory coordinate point sequences, speed change curves, and trigger time node parameters. By controlling environmental element changes in a time-sequential manner, the dynamic interference processes of adjacent vehicles cutting in and parking space obstruction in real-world scenarios are simulated.

[0104] Specifically, during the testing process, static scene parameters are used to initialize the parking space layout and obstacle positions, establishing a repeatable test baseline environment; parking direction parameters trigger the tested vehicle to perform a left or right turn to leave, verifying the directional adaptability of the system's path planning algorithm; and dynamic scene sequences, after the vehicle starts parking, drive the obstacle module to move according to preset time nodes and motion trajectories, forming a continuously changing test environment. These three elements work together to construct a complete testing framework from static to dynamic. The dynamic scene sequences, by precisely controlling the obstacle movement trajectory and triggering timing, can generate complex scenarios such as approaching vehicles accelerating and lateral movement obstructing traffic, thereby exposing decision-making vulnerabilities of the tested system in dynamic environments.

[0105] Compared to existing technologies, traditional manual testing can only set fixed obstacle positions and cannot adjust scene elements in real time during the test, resulting in a lack of dynamic environment simulation. This solution, however, achieves precise reproduction of obstacle movement trajectories through the temporal control of dynamic scene sequences, enabling the test scenario to cover unexpected interference situations that vehicles may encounter during parking. Furthermore, the standardized definition of static scene parameters and parking direction overcomes the problems of dimensional deviations and inconsistent operation timing that exist when manually setting up scenes.

[0106] Through the above technical solution, this application can systematically generate diverse test scenarios that include static layout and dynamic interference elements, accurately control the movement trajectory and triggering sequence of obstacles, and ensure that the execution conditions of different test cases are completely consistent, thereby effectively verifying the reliability of the vehicle's left and right parking functions in complex dynamic environments.

[0107] Next, refer to Figure 4 This invention describes another embodiment of an automated testing method for parking vehicles on the left and right.

[0108] Figure 4This is a flowchart illustrating the operation of an automated testing device for adjustable left and right parking scenarios of a vehicle, as provided in an embodiment of the present invention. The automated testing device for adjustable left and right parking scenarios of a vehicle employs the above-described... Figure 2 The test apparatus described herein is used in the host computer of the apparatus. Figure 4 The execution end includes a multi-degree-of-freedom dynamic scene generation system, a test control mobile phone, and a location sensing device.

[0109] The workflow of the automated testing device for adjustable parking scenarios for automobiles includes:

[0110] S1. Creating and Editing Test Cases: The operator first starts the host computer and creates or edits test cases in the integrated test control software. The test case content mainly includes: the operator editing the test script on the host computer, static scene parameters: setting the initial position (X, Y coordinates) and initial angle (R-axis rotation angle) of the left and right sides of the "multi-degree-of-freedom dynamic scene generation system" to accurately generate standard, non-standard, inclined, or narrow initial parking spaces. Parking task instructions: selecting the test task as "parking to the left" or "parking to the right". Dynamic scene sequence: programming to set the triggering conditions, timing, and motion parameters of dynamic events. For example: defining "1.5 seconds after the host computer sends the parking instruction, the obstacle on the left moves forward 1 meter along the Y-axis at a speed of 0.3 m / s".

[0111] S2. Execute automated scenario deployment and test startup: The operator clicks "Start Test" on the host computer. The host computer then starts executing the test script, first sending instructions to the lower-level motion controller to drive the slide rails and turntables in the multi-degree-of-freedom dynamic scene generation system, precisely moving and rotating the obstacles on the left and right sides to the initial position and posture set by the test case.

[0112] S3. After the scenario deployment is completed, the host computer sends preset ADB commands to the test control mobile phone via USB connection.

[0113] S4. Under ADB command control, test control mobile phone automatically simulates human hand clicking the "Park Left" or "Park Right" button on the mobile APP interface to accurately trigger the vehicle's parking function.

[0114] S5, the multi-degree-of-freedom dynamic scene generation system moves to the initial position and initial angle.

[0115] S6. Vehicle Parking Exit and Scene Dynamic Synchronization: After receiving the command from the mobile APP, the tested vehicle begins the parking exit operation. Simultaneously, the host computer, according to the synchronization logic set in the test cases, controls the multi-degree-of-freedom dynamic scene generation system to begin moving at a predetermined time, accurately simulating dynamic adversarial scenarios such as the simultaneous start-up of adjacent vehicles through dynamic changes. The tested vehicle responds in real-time to this dynamically changing environment during the parking exit process.

[0116] S7. Real-time monitoring and data acquisition: Throughout the entire parking process, the position sensing device (LiDAR / industrial camera) continuously monitors and records the real-time position, trajectory, and attitude data of the vehicle under test and dynamically moving obstacles, and transmits these high-precision position data streams back to the host computer in real time.

[0117] S8. The host computer software receives location data and performs real-time analysis: it calculates the minimum distance between the vehicle outline and the obstacle outline through an algorithm to determine whether a scratch or collision risk occurs during the parking process.

[0118] S9. After a parking test is completed, the host computer automatically determines the test result as "pass," "fail," or "collision" based on the vehicle's final stopping position and attitude, combined with preset success criteria. The host computer records the test result and then controls the vehicle to return to the initial parking space. Specifically, this can be achieved by sending an ADB command again to trigger the vehicle's "automatic parking" function, or by guiding the vehicle back along the planned path.

[0119] S10, Parking the vehicle in the parking space.

[0120] Once the vehicle has reset, a complete test cycle ends. At this point, the next round of testing or report generation can begin: the host computer can automatically load the next test case according to the test plan, repeating the steps to achieve large-scale, uninterrupted automated testing. After all test cases have been executed, the host computer will summarize the test results, process data, key screenshots, and other information from all cycles, automatically generating a detailed, structured comprehensive test report.

[0121] The testing method in this embodiment can systematically generate diverse test scenarios that include static layout and dynamic interference elements, accurately control the movement trajectory and triggering sequence of obstacles, ensure that the execution conditions of different test cases are completely consistent, solve the problem that dynamic scene changes are difficult to reproduce accurately, ensure the controllability and standardization of scene changes during the test, and provide a reliable dynamic test environment for verifying the decision logic of the automatic parking function.

[0122] Finally, it should be noted that the above technical solution is only one embodiment of the present invention. For those skilled in the art, based on the application methods and principles disclosed in the present invention, it is easy to make various types of improvements or modifications, and not limited to the methods described in the above specific embodiments of the present invention. Therefore, the methods described above are only preferred and have no limiting significance.

Claims

1. An automated testing device for adjustable left and right parking scenarios for automobiles, characterized in that, include: The test includes a host computer, a multi-degree-of-freedom dynamic scene generation system, a test control terminal, and a position sensing device. The host computer is communicatively connected to the multi-degree-of-freedom dynamic scene generation system, the test control terminal, and the position sensing device, respectively. The host computer for testing is used to respond to the user's test instructions and output a first control instruction to control the multi-degree-of-freedom dynamic scene generation system to generate the corresponding test scene. It also outputs a second control command to control the test control terminal to trigger the parking operation of the vehicle under test; The multi-degree-of-freedom dynamic scene generation system is used to receive the first control command and generate corresponding test scenes on the left and right sides of the vehicle under test based on the first control command. The test control terminal is used to receive the second control command and trigger the parking out or parking in operation of the vehicle under test based on the second control command. The position sensing device is used to collect the position data of the vehicle under test and the obstacle in real time, and send the collected position data to the test host computer; The host computer for testing is also used to receive the location data and generate test results based on the location data.

2. The automated testing device for adjustable left and right parking scenarios of automobiles according to claim 1, characterized in that: The multi-degree-of-freedom dynamic scene generation system includes a motion platform, which comprises: A two-dimensional moving mechanism is used to move along the X and Y axes in a horizontal plane to simulate the obstacle; A rotating turntable is positioned below the two-dimensional moving mechanism to support the two-dimensional moving mechanism and enable rotational adjustment. A motion controller is used to receive the first control command and control the movement of the two-dimensional moving mechanism and the rotating turntable.

3. The automated testing device for adjustable left and right parking scenarios of automobiles according to claim 2, characterized in that: The rotating turntable has a rotation angle range of 0 to 180 degrees and is used to simulate angled parking spaces or adjacent vehicles with incorrect posture.

4. The automated testing device for adjustable left and right parking scenarios of automobiles according to claim 1, characterized in that: The test control terminal is a smartphone, which communicates with the test host computer via the ADB protocol; the smartphone is used to receive ADB commands sent by the test host computer to automate the operation of the vehicle control APP.

5. The automated testing device for adjustable left and right parking scenarios of automobiles according to claim 1, characterized in that: The position sensing device includes a lidar and / or an image acquisition device, used to acquire the contour information and position information of the vehicle under test and the obstacle through point cloud analysis and / or image processing.

6. The automated testing device for adjustable left and right parking scenarios of automobiles according to claim 1, characterized in that: The host computer for testing is used for: Control the deployment of the initial parking space scene by the multi-degree-of-freedom dynamic scene generation system; During the parking process of the vehicle under test, the dynamic scene generation system is synchronously controlled to perform corresponding dynamic changes; Based on the data returned by the position sensing device, it is determined in real time whether the tested vehicle has collided or failed to park. After the test is completed, the vehicle under test is controlled to return to the initial parking space, and a test report is generated.

7. An automated testing method for automobiles parking left and right, characterized in that, Performed by the host computer of the apparatus as described in any one of claims 1 to 6, comprising: Receive test commands from users; In response to the test command, the multi-degree-of-freedom dynamic scene generation system is controlled to generate corresponding test scenes on the left and right sides of the vehicle under test by outputting a first control command; In response to the test command, the test control terminal is controlled by outputting a second control command to trigger the parking out or parking in operation of the vehicle under test; The system receives real-time position data of the vehicle and obstacles under test from the position sensing device and generates test results based on the position data.

8. The automated testing method for left and right parking of a vehicle according to claim 7, characterized in that: The test instructions include static scene parameters, departure direction, and dynamic scene sequence.

9. The automated testing method for left and right parking of a vehicle according to claim 8, characterized in that: The test scenarios are generated on the left and right sides of the vehicle under test, including: Based on the static scene parameters, the multi-degree-of-freedom dynamic scene generation system is controlled to deploy the initial parking space scene; During the parking process of the vehicle under test, the dynamic scene generation system is synchronously controlled to perform corresponding dynamic changes based on the dynamic scene sequence.

10. The automated testing method for left and right parking of a vehicle according to claim 8 or 9, characterized in that: The dynamic scene sequence includes controlling obstacles to move along a preset trajectory during the parking process of the vehicle under test, in order to simulate the dynamic behavior of the adjacent vehicle.

Images