A method for testing key parameters of a vehicle automatic parking system

Abstract

The present invention provides a method for testing key parameters of an automatic parking system of a vehicle, comprising: obtaining image data of the surrounding area of ​​the vehicle in a simulation scene through a pre-built simulation system; performing parking space detection and path planning through a preset automatic parking model based on the image data, generating control instructions, and controlling the vehicle to perform automatic parking based on the control instructions; setting different preset values ​​for key parameters during the automatic parking process, simulating parking effects under different preset values; and determining boundary warning values ​​of different key parameters according to the parking effects. The present invention solves the problem that the unclear basis for setting key parameters of the existing automatic parking system may lead to poor automatic parking effects.

Description

Technical Field

[0001] The present invention relates to the field of intelligent acceleration technology, and in particular to a method for testing key parameters of a vehicle automatic parking system. Background Art

[0002] As people's living standards improve, cars are becoming more and more popular. The continuous increase in urban construction facilities has increased traffic congestion while also reducing parking space. The narrow parking space increases the difficulty of parking, especially for inexperienced novices, and the demand for automatic parking assistance has greatly increased. Automatic parking assistance includes automatic parking search and automatic parking control functions. When searching for parking spaces, the environment perception components are used to identify and calculate the parking space information. The perception components include ultrasonic sensors, millimeter wave sensors, cameras, and laser sensors. During automatic parking control, the system controls the lateral and longitudinal movement of the vehicle according to the planned parking path to park the vehicle in place.

[0003] Based on the "Performance Requirements and Test Methods for Lane Departure Warning Systems of Intelligent Transport Systems" or "Test Procedures for Autonomous Driving Functions of Intelligent Connected Vehicles" (Trial) formulated by the International Organization for Standardization (ISO) or the National Standard of the People's Republic of China (GB / T), in a simulation environment, the safe working boundaries of the vision-based automatic parking system under the influence of key indicators such as latency / packet loss, quantification caused by resolution and bandwidth, perception output accuracy and variance are tested and analyzed, a target perception model based on the "Detection Method" is established, and quantitative analysis results of key indicators are obtained.

[0004] However, the safety working boundaries of existing vehicle automatic parking systems are affected by key indicators such as latency / packet loss, quantization caused by resolution and bandwidth, and perception output accuracy and variance, which affect automatic parking. It is difficult to accurately determine the parking position and measure the distance to surrounding objects, resulting in parking scratches or even failure to complete automatic parking. Summary of the invention

[0005] The present invention provides a method for testing key parameters of a vehicle automatic parking system, which is used to solve the problem that the key parameters of the existing automatic parking system are set based on unclear basis and may lead to poor automatic parking effect.

[0006] The present invention provides a method for testing key parameters of a vehicle automatic parking system, comprising:

[0007] Acquire image data around the vehicle in a simulation scenario through a pre-built simulation system;

[0008] Based on the image data, parking space detection and path planning are performed using a preset automatic parking model, a control instruction is generated, and the vehicle is controlled to perform automatic parking based on the control instruction;

[0009] During the automatic parking process, different preset values ​​are set for key parameters to simulate the parking effects under different preset values;

[0010] Boundary warning values ​​of different key parameters are determined according to the parking effect.

[0011] According to a method for testing key parameters of an automatic parking system of a vehicle provided by the present invention, the method of acquiring image data of the surrounding area of ​​the vehicle in a simulation scene by means of a pre-built simulation system specifically includes:

[0012] The vehicle-mounted fisheye camera is used to obtain image data of the vehicle's surroundings and perform distortion correction and perspective conversion processing to generate a preprocessed image.

[0013] According to a method for testing key parameters of a vehicle automatic parking system provided by the present invention, parking space detection and path planning are performed based on the image data through a preset automatic parking model, a control instruction is generated, and the vehicle is controlled to perform automatic parking based on the control instruction, specifically comprising:

[0014] Based on the pre-processed image, the parking space is detected by using a preset automatic parking model to identify the parking space;

[0015] Based on the identified parking spaces, the parking path is planned using the arc and straight line method;

[0016] A control command is generated based on the path planning scheme, and the steering wheel angle and vehicle speed are controlled by the control command to perform automatic parking.

[0017] According to a method for testing key parameters of a vehicle automatic parking system provided by the present invention, different preset values ​​are set for key parameters during the automatic parking process to simulate parking effects under different preset values, including:

[0018] The resolution of the key parameters is set to a fixed value, the frame rate is changed and different preset values ​​are set for the time delay and the packet loss rate, and the left distance between the left side of the vehicle body and the left side of the storage edge is simulated and calculated;

[0019] Among them, the key parameters include: time delay, packet loss rate, resolution and perception output accuracy.

[0020] According to a method for testing key parameters of a vehicle automatic parking system provided by the present invention, different preset values ​​are set for key parameters during the automatic parking process to simulate parking effects under different preset values, and further includes:

[0021] The time delay and packet loss rate in the key parameters are set to fixed values, the frame rate is changed and different preset values ​​are set for the resolution in the key parameters, and the left side distance between the left side of the vehicle body and the left side of the library is simulated and calculated.

[0022] According to a method for testing key parameters of a vehicle automatic parking system provided by the present invention, different preset values ​​are set for key parameters during the automatic parking process to simulate parking effects under different preset values, and further includes:

[0023] Set a fixed frame rate, change the resolution and set the perception output accuracy to different preset values, and simulate the calculation of the left distance between the left side of the vehicle body and the left side of the library.

[0024] According to a method for testing key parameters of a vehicle automatic parking system provided by the present invention, determining boundary warning values ​​of different key parameters according to the parking effect specifically includes:

[0025] According to the comparison between the left distance and the set threshold, the boundary warning values ​​of different key parameters are determined;

[0026] When the left distance is greater than the first set threshold, it means that the vehicle has not crossed the line during the parking process;

[0027] The left distance is between the first set threshold and zero, indicating that the vehicle is crossing the line during the parking process;

[0028] When the left distance is less than zero or greater than a second set threshold, it indicates that the vehicle has not entered the warehouse.

[0029] The present invention also provides a vehicle automatic parking system key parameter testing system, the system comprising:

[0030] A data acquisition module, used to acquire image data of the vehicle's surroundings in a simulation scene through a pre-built simulation system;

[0031] An automatic parking module, configured to perform parking space detection and path planning based on the image data through a preset automatic parking model, generate control instructions, and control the vehicle to perform automatic parking based on the control instructions;

[0032] A simulation module, used to set different preset values ​​for key parameters during the automatic parking process and simulate parking effects under different preset values;

[0033] The boundary warning value determination module determines the boundary warning values ​​of different key parameters according to the parking effect.

[0034] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, a key parameter testing method for an automatic parking system for a vehicle as described in any one of the above is implemented.

[0035] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the method for testing key parameters of a vehicle automatic parking system as described in any one of the above is implemented.

[0036] The present invention provides a method for testing key parameters of a vehicle automatic parking system. Different key parameters are tested under preset values ​​through a simulation system. Boundary warning values ​​of the key parameters are found through testing. Based on the boundary warning values, the automatic parking model can be better adjusted to perform automatic parking, thereby improving accuracy, avoiding scratches, and achieving safe parking. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0038] Figure 1 It is a flow chart of a method for testing key parameters of a vehicle automatic parking system provided by the present invention;

[0039] Figure 2 It is a schematic diagram of vertical parking path planning provided by the present invention;

[0040] Figure 3 (a) is a schematic diagram of a single storage provided by the present invention;

[0041] Figure 3 (b) is a schematic diagram of the parking angle during the parking process provided by the present invention;

[0042] Figure 3 (c) is a schematic diagram of the parking introduction provided by the present invention;

[0043] Figure 4 It is the left distance analysis diagram under different time delays provided by the present invention;

[0044] Figure 5 It is a schematic diagram of a Markov chain packet loss model with two states provided by the present invention;

[0045] Figure 6 It is a left distance analysis diagram under different packet loss rates provided by the present invention;

[0046] Figure 7 The left distance analysis diagram under different sampling frequencies and resolutions provided by the present invention is

[0047] Figure 8It is a module connection diagram of a key parameter test system of a vehicle automatic parking system provided by the present invention;

[0048] Fig. 9 It is a structural schematic diagram of the electronic device provided by the present invention.

[0049] Reference numerals:

[0050] 110: data acquisition module; 120: automatic parking module; 130: simulation module; 140: boundary warning value determination module;

[0051] 910: processor; 920: communication interface; 930: memory; 940: communication bus. DETAILED DESCRIPTION

[0052] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0053] Combine the following Figure 1 A method for testing key parameters of an automatic parking system of a vehicle according to the present invention is described, comprising:

[0054] S100, acquiring image data of the surrounding area of ​​the vehicle in a simulation scene through a pre-built simulation system;

[0055] In the present invention, the video image input by the vehicle-mounted fisheye camera is an M×N×3 color image array, also known as an RGB image. Since the imaging characteristics of the fisheye camera will cause obvious distortion of the captured image, it is necessary to first perform distortion correction on the fisheye camera, and then use the top view transformation algorithm to convert the corrected distortion-free image perspective into a top view.

[0056] The Zhang Zhengyou calibration method is used to correct the distortion of the fisheye camera. The camera takes multiple chessboard images. All corners of the chessboard must be photographed, and the calibration plates are distributed as evenly as possible in various areas of the picture.

[0057] The fisheye camera is automatically calibrated using the Camera Calibrator program in MATLAB to obtain the internal and external parameters of the matrix. Finally, the image distortion is corrected based on the internal and external parameters of the camera obtained through calibration. Although the corrected image is not as clear as before, the distortion has been basically eliminated, and the result can meet the requirements of subsequent designs.

[0058] Then, we transform the top view. When installing the fisheye camera, there will be a certain angle between the optical axis of the camera and the ground. The image corrected in the previous section still cannot directly reflect the real situation on the ground, so we should transform the corrected image into a top view. According to the internal and external parameters of the fisheye camera, the inverse projection transformation algorithm is used to transform the corrected image into a top view image, and the area of ​​interest is cropped. Distortion correction and perspective conversion processing will help to identify the parking space more quickly and accurately in the future, creating a real and accurate environment for simulation.

[0059] S200, performing parking space detection and path planning based on the image data using a preset automatic parking model, generating a control instruction, and controlling the vehicle to perform automatic parking based on the control instruction;

[0060] In the present invention, parking space monitoring includes the following steps: the image captured by the camera is a color image based on the RGB color mode, and the color of each pixel can be expressed by formula (1), that is, the three components R, G, B (red, green, blue) together determine the color of the pixel.

[0061] C(x,y)=[R(x,y),G(x,y),B(x,y)] T =[R,G,B] T (1)

[0062] Among them, (x, y) represents the coordinates of the pixel point. This section deals with the image of the parking angle area. This type of image is also an RGB color image, which contains a large amount of data and has low processing efficiency. In computer vision, color images are usually grayed out first, which can reduce the amount of calculation while retaining the main information of the image. Graying is to make the three values ​​of R, G, and B of the image equal through certain calculations. This equal value is the gray value. Generally speaking, there are four algorithms for graying images: component method, maximum value method, average method, and weighted average method.

[0063] During the process of collecting images with a camera, due to the camera's own characteristics and changes in external conditions, the original panoramic image collected will have certain noise, which will affect the subsequent image processing, so image filtering should be performed first to weaken the interference of noise. Common filtering methods include mean filtering, median filtering and Gaussian filtering.

[0064] Threshold segmentation is performed. Compared with the ground, the parking space line has a higher degree of recognition and a clear color contrast. Therefore, the image can be directly binarized to filter the non-parking space line area and obtain the parking space line area. Image binarization is based on the set threshold. The grayscale value of pixels below this threshold is 0, and the grayscale value of pixels above this threshold is 255, thus presenting a black and white contrast image. Its expression is as follows:

[0065]

[0066] Where T is the threshold. The key to binarization is the determination of the threshold, which determines the final processing effect. If the threshold is too small, some irrelevant background information may be retained; if the threshold is too large, some parking line information may be lost, seriously affecting the subsequent image processing. The binarization algorithm has two types: fixed threshold method and adaptive threshold method. The fixed threshold method is suitable for processing a single image, but here it needs to be applicable to a large number of parking line images, so this paper uses the adaptive threshold method to binarize the parking line image.

[0067] To perform skeleton monitoring, after binarizing the parking space angle image, it is necessary to extract the skeleton of the parking space line on the processed image. The skeleton is a compact representation of a shape and can be regarded as a stable central axis equidistant from the shape boundary. It has the same topological structure and geometric properties as a two-dimensional shape. The skeleton also contains the contour information and area information of the two-dimensional shape, so we can use the skeleton to represent the target shape. The traditional distance transform method is used to extract the skeleton, which has obvious advantages in the accuracy of the skeleton points. The skeleton is extracted from the binarized corrected parking space image.

[0068] Horizontal straight line detection is performed, and the straight line in the skeleton image is extracted using Hough transform to identify the skeleton straight line of the parking space line. In the image space (x, y), the points on the same straight line satisfy the equation y=kx+b, that is, the slope k of the points on the same straight line is equal and the intercept is also equal. By constructing the parameter space (k, b), it can be concluded that a straight line in the image space (x, y) corresponds to a point in the parameter space (k, b). Using the point-line duality between the parameter space and the image space, the Hough transform converts the straight line detection problem in the image space into the parameter space: the cumulative statistics in the parameter space (k, b) are performed to find the counter peaks of the slope and intercept to detect the straight line in the image space. The horizontal straight line is screened out using the slope of the straight line, which is the parking space line. The intersection point (xr, yr) of the vertical center line of the image and the parking space line is extracted in real time.

[0069] Mark the parking space line information, mark the extracted intersection (xr, yr) in the top view, and use MATLAB's imageToVehicle function to convert the coordinates in the top view to the vehicle body coordinate system to calculate the relative distance between the parking space line and the vehicle body.

[0070] To identify parking spaces, the visual sensors on both sides of the vehicle body detect the distance between the vehicle and the parking space line in real time. When the vehicle passes through the parking space, the distance between the vehicle and the parking space line detected by the visual sensor on the side of the vehicle body will jump twice. The reason for the two jumps is that the visual sensor detects the boundary line of the parking space. If the position of the visual sensor at each jump can be recorded, the position of the left and right boundary lines of the parking space will be determined accordingly. The vehicle drives along the positive direction of the X-axis, and the visual sensor continuously detects the surrounding situation and generates the parking space line image on both sides of the vehicle in real time. The simulation results show that the visual sensor can effectively identify the parking space, and can obtain the location and length and width information of the parking space to realize parking space recognition.

[0071] After completing the parking space identification, the path planning is carried out, which includes:

[0072] The arc straight line method is used to plan the vertical parking path. The vehicle enters the garage once, that is, the vehicle goes from the starting point position P1 to the garage P4 through a reverse movement, without stopping or changing direction in the middle. Figure 2 As shown in the figure, the vehicle starts from point P1, first goes a straight line to point P2, then enters the arc section (minimum turning radius), and when the vehicle heading angle is parallel to the parking space, it backs up to the target parking space in a straight line to complete a parking. Point P1 is the starting position of the vehicle, which is collected by the GPS sensor. Point P2 is the tangent point of the arc and the horizontal line, which is generated by the path planning module. Point P3 is the tangent point of the arc and the vertical line, which is generated by the parking space detection module. The main advantage of this path planning method is that the steering wheel angle can be easily obtained. For vertical parking control, the steering wheel angle is 0 when driving in a straight line, and the arc steering wheel angle is + / - the maximum value. During the transition process, the steering wheel angle increases or decreases linearly, which is to prevent the vehicle from shaking. The vehicle speed is set to 10km / h when searching for a parking space and 3.6km / h when reversing into the garage.

[0073] In addition, after the path planning is completed in the present invention, a human-computer interaction process is also provided. The driver decides whether to start the parking command by interacting with the vehicle, and can start parking by voice or button. The parking process of the vehicle is displayed in real time on the human-computer interaction platform, and if a situation occurs during the parking process, the parking information will be fed back to the driver. The present invention realizes the research of automatic parking algorithm based on vision based on Matlab / Simulink and Prescan joint simulation.

[0074] The vision-based automatic parking system of the present invention can identify the vacant parking space and judge the parking space type and size through the camera by the environment perception module composed of the vehicle's sensors without the intervention of the driver. During the parking process, the ultrasonic radar will also monitor the surrounding obstacles at all times. Then the information and data collected by the environment perception module are transmitted to the planning and decision-making module through communication. The controller responsible for the planning and decision-making function will process and calculate the information transmitted by the sensor. The parking controller will judge whether the detected target parking space can be parked. If it can be parked, different parking paths will be selected for parking according to the type of parking space and the parking environment. Finally, the steering wheel angle and vehicle speed determined by the decision are transmitted to the actuator in real time, so that the actuator controls the movement of the vehicle and allows the vehicle to park safely and quickly in the target parking space according to the planned trajectory. The automatic parking system also has a human-computer interaction process. The driver decides whether to start the parking command by interacting with the vehicle, and can start parking by voice or button. The parking process of the vehicle is displayed in real time on the human-computer interaction platform, and the parking information will be fed back to the driver if there is a situation during the parking process. When the vehicle is parked safely and smoothly in the parking space, the interactive platform will send parking completion information to the driver.

[0075] S300, setting different preset values ​​for key parameters during the automatic parking process to simulate parking effects under different preset values;

[0076] In the present invention, vertical parking is used as a simulation test scenario. Since the distance between the vehicle body and the parking lot edge (i.e., y2) has been determined, combined with literature data and empirical analysis, whether the vehicle can be successfully parked depends on the x-axis position of the vehicle (i.e., x2) when the steering wheel angle just starts to change during the reversing process. Figure 3 As shown in (a). Since x2 cannot be measured directly, the distance between the left side of the vehicle body and the left side of the parking garage after entering the parking garage (left side distance) is selected as the evaluation index, as shown in Figure 3 When the left distance is greater than 1.024m, it means that the vehicle has not crossed the line during the parking process; when the left distance is less than 1.024m, it means that the vehicle has crossed the line during the parking process; when the left distance is less than 0 or greater than 1.990m, it means that the vehicle has not entered the parking lot.

[0077] For the delay test analysis, in the unreliable communication scenario, the distance between the vehicle body and the parking space line is detected by the camera, and there is a delay in the position. The delay will reduce the performance of the system. For the delay under unreliable communication, this test uses the delay module of the Simulink library, sets the camera resolution to 1280×960, and sets different frame rates and time delays for the test. The data is shown in Table 1:

[0078] Table 1 Data records of left distance under different frame rates and delays

[0079]

[0080] From Table 1, we can see that the impact of delay on APS is significant. As the delay increases, its impact increases. The three-dimensional analysis diagram of delay, frame rate and distance on the left is shown in Figure 1. Figure 4 As shown. Figure 4 The red boundary line in the figure is the failure boundary of the automatic parking system under delay. As can be seen from the figure, with the increase of delay, the distance on the left side drops sharply. When the frame rate is 25fps and the delay is greater than 50ms, the vehicle crosses the line; when the frame rate is 30, 40 and 60fps, the vehicle crosses the line only when the delay is greater than 100ms. Because when the delay is a constant value, the vehicle tracks the historical state of the parking space line before the delay time length in the steady state, so there is a lag phenomenon in the parking space. The parking space lag distance is the distance traveled by the vehicle under the delay time length. Due to the delay, the parking space position received by the automatic parking system lags behind the actual parking space position. Therefore, the actual parking position of the vehicle lags behind the theoretical parking position, causing the distance on the left side to continue to decrease, so the vehicle crosses the line. The experimental results show that delay is one of the key factors affecting the performance of APS.

[0081] For packet loss rate test analysis, a packet loss model is built in Simulink. In communication networks, data transmission is usually bursty, that is, after a data is lost during transmission, the probability of the next data being lost is greater than the probability of successful transmission. This characteristic is usually expressed by a Figure 5 It is described by a Markov chain with two states as shown in the figure. Where σ(k) = 1 means that the data is not lost when it is transmitted through the network, and the system is in a closed loop state; σ(k) = 2 means that the data is lost when it is transmitted through the network, and the system is in an open loop state. The state transition probability matrix of the Markov chain is:

[0082] P = [p ij ],

[0083] p ij =P{σ(k+1)=j|σ(k)=i},

[0084]

[0085] Set different packet loss probabilities to 5%, 25%, 35%, 55%, 65%, and 80% for detection tests. Set the camera's individual resolution to 1280×960, and the left distances under different frame rates and packet loss rates are shown in Table 2 below:

[0086] Table 2 Data records of left distance under different frame rates and packet loss rates

[0087]

[0088] As can be seen from Table 2, as the packet loss rate increases, its impact increases accordingly. The three-dimensional analysis diagram of packet loss rate, frame rate and distance on the left is shown in Figure 6 As shown. Figure 6 Packet loss data analysis shows that when the packet loss rate is lower than 55% in the steady state, the rate of decrease of the left distance with the packet loss rate is very small, and the packet loss is regarded as the increase of the input signal sampling period, while the closed-loop control period remains unchanged. Since the control system has the properties of a low-pass filter, the controller has a certain robustness to changes in the input signal sampling period, which is manifested as a certain resistance to occasional packet loss and slight degradation of control performance. When the packet loss rate is higher than 55%, the left distance decreases rapidly. When the packet loss rate is higher than 65%, the parking space detection module cannot correctly search for available parking spaces and the vehicle crosses the line. Packet loss when the distance between the vehicle body and the parking space line changes rapidly, especially continuous packet loss, has a significant impact on the control error. In this test, the packet loss rate of 65% can be regarded as a turning point, and the actual value should be related to the system characteristics and working conditions.

[0089] The quantitative test analysis caused by resolution is carried out. The camera is used to detect the distance between the vehicle body and the parking space line, so the camera resolution will affect the parking performance of the system. Considering the resolution size of the camera in actual applications, 5 groups of different resolutions are taken, and the parking performance of APS is tested at each resolution. Table 3 records the left distance between the two vehicles under different experiments.

[0090] Table 3 Data records of left distance at different resolutions and frame rates

[0091]

[0092] From Table 3, we can see that the left distance generally increases with the increase of resolution. The three-dimensional analysis diagram of resolution, frame rate and left distance is shown in Figure 7 shown.

[0093] According to the perception output accuracy test analysis, the accuracy of the relative distance between the vehicle body and the parking space line detected by the camera in the system will directly affect the output accuracy of the system. The present invention uses MATLAB Function to generate the actual relative distance between the vehicle body and the parking space line in real time. However, the relative distance measured by the vision-based distance measurement method in this paper has a certain deviation from the actual relative distance. The frame rate of the camera is set to 30fps, and the absolute error mean, variance and left distance data of the actual relative distance and the measured relative distance under different perception output accuracies are given, as shown in Table 4.

[0094] Table 4 Mean, variance and left distance of absolute error at different resolutions

[0095]

[0096] It can be seen from Table 4 that, under different resolutions, the larger the absolute error mean and variance, the shorter the left distance, and the easier it is for the vehicle to cross the line. When the absolute error mean and variance of the perception ranging module are greater than 0.3684m and 1.6651m2 respectively, the vehicle crosses the parking corner.

[0097] S400: Determine boundary warning values ​​of different key parameters according to the parking effect.

[0098] According to the comparison between the left distance and the set threshold, the boundary warning values ​​of different key parameters are determined;

[0099] When the left distance is greater than the first set threshold, it means that the vehicle has not crossed the line during the parking process;

[0100] The left distance is between the first set threshold and zero, indicating that the vehicle is crossing the line during the parking process;

[0101] When the left distance is less than zero or greater than a second set threshold, it indicates that the vehicle has not entered the warehouse.

[0102] Through the key parameter testing method of the vehicle automatic parking system provided by the present invention, different key parameters are tested under preset values ​​through a simulation system, and the boundary warning values ​​of the key parameters are found through testing. Based on the boundary warning values, the automatic parking model can be better adjusted to perform automatic parking, improve accuracy, avoid the occurrence of scratches, and achieve safe parking.

[0103] refer to Figure 8 The present invention also discloses a key parameter testing system for a vehicle automatic parking system, the system comprising:

[0104] A data acquisition module 110, for acquiring image data of the surroundings of the vehicle in a simulation scene through a pre-built simulation system;

[0105] The automatic parking module 120 is used to perform parking space detection and path planning based on the image data through a preset automatic parking model, generate a control instruction, and control the vehicle to perform automatic parking based on the control instruction;

[0106] A simulation module 130, used to set different preset values ​​for key parameters during the automatic parking process, and simulate parking effects under different preset values;

[0107] The boundary warning value determination module 140 determines boundary warning values ​​of different key parameters according to the parking effect.

[0108] Among them, the data acquisition module obtains the vehicle's surrounding image data through the on-board fisheye camera and performs distortion correction and perspective conversion processing to generate a pre-processed image.

[0109] The automatic parking module detects parking spaces based on the pre-processed images and uses a preset automatic parking model to identify parking spaces;

[0110] Based on the identified parking spaces, the parking path is planned using the arc and straight line method;

[0111] A control command is generated based on the path planning scheme, and the steering wheel angle and vehicle speed are controlled by the control command to perform automatic parking.

[0112] A simulation module, wherein the resolution among the key parameters is set to a fixed value, the frame rate is changed and different preset values ​​are set for the time delay and the packet loss rate, and the left distance between the left side of the vehicle body and the left side of the storage area is simulated and calculated;

[0113] Among them, the key parameters include: time delay, packet loss rate, resolution and perception output accuracy.

[0114] The time delay and packet loss rate in the key parameters are set to fixed values, the frame rate is changed and different preset values ​​are set for the resolution in the key parameters, and the left side distance between the left side of the vehicle body and the left side of the library is simulated and calculated.

[0115] Set a fixed frame rate, change the resolution and set the perception output accuracy to different preset values, and simulate the calculation of the left distance between the left side of the vehicle body and the left side of the library.

[0116] The boundary warning value determination module determines the boundary warning values ​​of different key parameters based on the comparison between the left distance and the set threshold;

[0117] When the left distance is greater than the first set threshold, it means that the vehicle has not crossed the line during the parking process;

[0118] The left distance is between the first set threshold and zero, indicating that the vehicle is crossing the line during the parking process;

[0119] When the left distance is less than zero or greater than a second set threshold, it indicates that the vehicle has not entered the warehouse.

[0120] A vehicle automatic parking system key parameter testing system provided by the present invention tests different key parameters under preset values ​​through a simulation system, and finds the boundary warning values ​​of the key parameters through testing. Based on the boundary warning values, the automatic parking model can be better adjusted to perform automatic parking, thereby improving accuracy, avoiding scratches, and achieving safe parking.

[0121] Fig. 9 An example of a physical structure diagram of an electronic device is shown in FIG. Fig. 9As shown, the electronic device may include: a processor 910, a communication interface 920, a memory 930 and a communication bus 940, wherein the processor 910, the communication interface 920 and the memory 930 communicate with each other through the communication bus 940. The processor 910 may call the logic instructions in the memory 930 to execute a method for testing key parameters of a vehicle automatic parking system, the method comprising: acquiring image data of the vehicle periphery in a simulation scene through a pre-built simulation system;

[0122] Based on the image data, parking space detection and path planning are performed using a preset automatic parking model, a control instruction is generated, and the vehicle is controlled to perform automatic parking based on the control instruction;

[0123] During the automatic parking process, different preset values ​​are set for key parameters to simulate the parking effects under different preset values;

[0124] Boundary warning values ​​of different key parameters are determined according to the parking effect.

[0125] In addition, the logic instructions in the above-mentioned memory 930 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a 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, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0126] On the other hand, the present invention further provides a computer program product, the computer program product includes a computer program, the computer program can be stored in a non-transitory computer-readable storage medium, when the computer program is executed by a processor, the computer can execute a vehicle automatic parking system key parameter testing method provided by the above methods, the method comprising: acquiring image data of the vehicle surroundings in a simulation scene through a pre-built simulation system;

[0127] Based on the image data, parking space detection and path planning are performed using a preset automatic parking model, a control instruction is generated, and the vehicle is controlled to perform automatic parking based on the control instruction;

[0128] During the automatic parking process, different preset values ​​are set for key parameters to simulate the parking effects under different preset values;

[0129] Boundary warning values ​​of different key parameters are determined according to the parking effect.

[0130] In another aspect, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, a method for testing key parameters of an automatic parking system of a vehicle provided by the above methods is implemented, the method comprising: acquiring image data of the surrounding area of ​​the vehicle in a simulation scene through a pre-built simulation system;

[0131] Based on the image data, parking space detection and path planning are performed using a preset automatic parking model, a control instruction is generated, and the vehicle is controlled to perform automatic parking based on the control instruction;

[0132] During the automatic parking process, different preset values ​​are set for key parameters to simulate the parking effects under different preset values;

[0133] Boundary warning values ​​of different key parameters are determined according to the parking effect.

[0134] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0135] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0136] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for testing key parameters of a vehicle automatic parking system, characterized in that: include: Acquire image data around the vehicle in a simulation scenario through a pre-built simulation system; Based on the image data, parking space detection and path planning are performed using a preset automatic parking model, a control instruction is generated, and the vehicle is controlled to perform automatic parking based on the control instruction; During the automatic parking process, different preset values ​​are set for key parameters to simulate the parking effects under different preset values; Determining boundary warning values ​​of different key parameters according to the parking effect; The method of setting different preset values ​​for key parameters during the automatic parking process to simulate parking effects under different preset values ​​includes: The resolution in the key parameters is set to a fixed value, the frame rate is changed and different preset values ​​are set for the time delay and the packet loss rate, and the left distance between the left side of the vehicle body and the left side of the library is simulated and calculated; the key parameters include: time delay, packet loss rate, resolution and perception output accuracy; The time delay and packet loss rate in the key parameters are set to fixed values, the frame rate is changed and different preset values ​​are set for the resolution in the key parameters, and the left distance between the left side of the vehicle body and the left side of the library is simulated and calculated; Set a fixed frame rate, change the resolution and set the perception output accuracy to different preset values, and simulate the calculation of the left distance between the left side of the vehicle body and the left side of the library.

2. A method for testing key parameters of a vehicle automatic parking system according to claim 1, characterized in that: The method of obtaining image data of the surrounding area of ​​the vehicle in the simulation scene by using the pre-built simulation system specifically includes: The vehicle-mounted fisheye camera is used to obtain image data of the vehicle's surroundings and perform distortion correction and perspective conversion processing to generate a preprocessed image.

3. The method for testing key parameters of a vehicle automatic parking system according to claim 1, characterized in that: Based on the image data, parking space detection and path planning are performed through a preset automatic parking model, and a control instruction is generated. Based on the control instruction, the vehicle is controlled to perform automatic parking, specifically including: Based on the pre-processed image, the parking space is detected by using a preset automatic parking model to identify the parking space; Based on the identified parking spaces, the parking path is planned using the arc and straight line method; A control command is generated based on the path planning scheme, and the steering wheel angle and vehicle speed are controlled by the control command to perform automatic parking.

4. The method for testing key parameters of a vehicle automatic parking system according to claim 1, characterized in that: Determining the boundary warning values ​​of different key parameters according to the parking effect specifically includes: According to the comparison between the left distance and the set threshold, the boundary warning values ​​of different key parameters are determined; When the left distance is greater than the first set threshold, it means that the vehicle has not crossed the line during the parking process; The left distance is between the first set threshold and zero, indicating that the vehicle is crossing the line during the parking process; When the left distance is less than zero or greater than a second set threshold, it indicates that the vehicle has not entered the warehouse.

5. A vehicle automatic parking system key parameter testing system, characterized in that: The system comprises: A data acquisition module, used to acquire image data of the vehicle's surroundings in a simulation scene through a pre-built simulation system; An automatic parking module, configured to perform parking space detection and path planning based on the image data through a preset automatic parking model, generate control instructions, and control the vehicle to perform automatic parking based on the control instructions; A simulation module, used to set different preset values ​​for key parameters during the automatic parking process and simulate parking effects under different preset values; A boundary warning value determination module, which determines boundary warning values ​​of different key parameters according to the parking effect; The method of setting different preset values ​​for key parameters during the automatic parking process to simulate parking effects under different preset values ​​includes: The resolution in the key parameters is set to a fixed value, the frame rate is changed and different preset values ​​are set for the time delay and the packet loss rate, and the left distance between the left side of the vehicle body and the left side of the library is simulated and calculated; the key parameters include: time delay, packet loss rate, resolution and perception output accuracy; The time delay and packet loss rate in the key parameters are set to fixed values, the frame rate is changed and different preset values ​​are set for the resolution in the key parameters, and the left distance between the left side of the vehicle body and the left side of the library is simulated and calculated; Set a fixed frame rate, change the resolution and set the perception output accuracy to different preset values, and simulate the calculation of the left distance between the left side of the vehicle body and the left side of the library.

6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the key parameter testing method of the vehicle automatic parking system as claimed in any one of claims 1 to 4 is implemented.

7. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, a method for testing key parameters of an automatic parking system for a vehicle as claimed in any one of claims 1 to 4 is implemented.

Images