Vehicle and parking control method and device thereof

By combining the vehicle panoramic image and positioning data of the Ackerman steering model in the automatic parking system, using the Kalman filter model to fusion data, the problem of insufficient vehicle positioning accuracy in the prior art is solved, and higher parking control is achieved Accuracy and wider application.

CN118907074APending Publication Date: 2024-11-08DONGFENG MOTOR GRP
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411036483.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-31
Publication Date
2024-11-08

AI Technical Summary

Technical Problem

The existing automatic parking system has insufficient accuracy and application limitations in vehicle positioning, especially under the influence of light, shadow and occlusion, which leads to the failed tracking of the target parking space marking points and the inability to accurately locate.

Method used

By receiving parking task instructions, combining the vehicle panoramic image and the first positioning data determined by the Ackerman steering model, the second positioning data is obtained, and the Kalman filter model is used to fuse the two positioning data to accurately control the vehicle to perform parking tasks and make up for it. Ackman steering model cumulative errors in the middle and late stages of parking tasks.

Benefits of technology

It improves the accuracy of vehicle automatic parking control, reduces dependence on other sensors, realizes large-scale automatic parking applications, and reduces application limitations.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118907074A_ABST
    Figure CN118907074A_ABST
Patent Text Reader

Abstract

The invention discloses a vehicle and a parking control method and device thereof, and relates to the technical field of parking control. According to the technical scheme, after the execution instruction of the parking task is received, the target parking space of the to-be-parked vehicle is determined according to the execution instruction, the vehicle panoramic image of the to-be-parked vehicle and the first positioning data determined based on the Ackerman steering model are obtained, and the second positioning data are obtained according to the vehicle panoramic image; the first positioning data and the second positioning data are data of the relative position relation between the to-be-parked vehicle and the target parking space, so that the to-be-parked vehicle can be accurately controlled to execute the parking task according to the first positioning data and the second positioning data until the to-be-parked vehicle moves into the target parking space. According to the technical scheme, parking is performed by fusing the first positioning data and the second positioning data, so that the vehicle panoramic image effectively compensates for accumulated errors of the Ackerman steering model in the middle and later periods of a parking task, and the accuracy of automatic parking control of the vehicle is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of parking control, and in particular to a vehicle and a parking control method and device thereof. Background Art

[0002] As an effective way to solve the problem of parking difficulties, automatic parking technology has gradually become popular in ordinary car models in recent years. At present, the automatic parking system mainly includes modules such as detection and tracking of parking spaces and obstacles, vehicle positioning, parking trajectory planning and vehicle control. Among them, the vehicle positioning of automatic parking is a prerequisite for the automatic parking system to perform path planning and vehicle control. At present, most of the vehicles equipped with automatic parking systems on the market use high-precision sensors such as GPS (Global Positioning System), IMU (Inertial Measurement Unit), LiDAR, etc. to achieve vehicle positioning, resulting in limitations in the promotion and application of automatic parking. Summary of the invention

[0003] In view of the above technical problems, the present invention provides a vehicle and a parking control method and device thereof, which improve the accuracy of vehicle automatic parking control and reduce the limitations of vehicle automatic parking applications.

[0004] The embodiment of the present invention provides the following solution:

[0005] In a first aspect, an embodiment of the present invention provides a parking control method, the method comprising:

[0006] Receiving execution instructions of parking tasks;

[0007] Determine a target parking space for the vehicle to be parked according to the execution instruction, and obtain a panoramic image of the vehicle to be parked and first positioning data determined based on an Ackerman steering model;

[0008] Obtaining second positioning data according to the panoramic image of the vehicle, wherein the first positioning data and the second positioning data are both data of the relative position relationship between the to-be-parked vehicle and the target parking space;

[0009] According to the first positioning data and the second positioning data, the vehicle to be parked is controlled to perform the parking task until the vehicle to be parked moves into the target parking space.

[0010] In an optional embodiment, obtaining first positioning data determined by the vehicle to be parked based on the Ackerman steering model includes:

[0011] Processing the steering data of the vehicle to be parked based on the Ackerman steering model to obtain a first heading angle of the ground coordinate system of the vehicle to be parked at the target parking space in different positioning cycles;

[0012] Obtaining a first coordinate value of the rear axle center of the vehicle to be parked in each positioning cycle in the ground coordinate system according to the first heading angle of each positioning cycle and the pulse data collected by the wheel speed sensor on the vehicle to be parked;

[0013] The first positioning data is output according to the first coordinate value.

[0014] In an optional embodiment, obtaining the first coordinate value of the rear axle center of the vehicle to be parked in each positioning cycle in the ground coordinate system according to the first heading angle of each positioning cycle and the pulse data collected by the wheel speed sensor on the vehicle to be parked includes:

[0015] According to the formula:

[0016]

[0017] Obtain the horizontal coordinate increment Δx and vertical coordinate increment Δy of the axis center in the ground coordinate system after adjacent positioning cycles, where BL n and BR n are the number of left rear wheel pulses and right rear wheel pulses of the vehicle to be parked in adjacent positioning cycles, Δd is the wheel travel distance of a single pulse, Δθ is the first heading angle increment in adjacent positioning cycles, and θ t is the first heading angle of the previous positioning cycle;

[0018] The first coordinate value of the vehicle to be parked in each positioning cycle is obtained according to the abscissa increments and ordinate increments of adjacent positioning cycles and the abscissa and ordinate of the vehicle to be parked in the previous positioning cycle.

[0019] In an optional embodiment, obtaining second positioning data according to the vehicle panoramic image includes:

[0020] Obtaining the center coordinate value of the rear axle of the vehicle to be parked in the parking space coordinate system according to the vehicle panoramic image, wherein the parking space coordinate system is a coordinate system constructed based on the location of the target parking space on the vehicle panoramic image;

[0021] According to the rear axle center coordinate value and a preset coordinate mapping relationship, a second coordinate value of the vehicle to be parked in the ground coordinate system is obtained, wherein the coordinate mapping relationship is a position mapping relationship between the parking space coordinate system and the ground coordinate system;

[0022] The second positioning data is output according to the second coordinate value.

[0023] In an optional embodiment, before the vehicle to be parked is controlled to perform the parking task according to the first positioning data and the second positioning data, the method further includes:

[0024] According to the timestamp of the second positioning data, determining the first positioning data having the same timestamp in the data queue of the first positioning data;

[0025] The first positioning data and the second positioning data with the same timestamp are determined as target data for the vehicle to be parked to perform the parking task.

[0026] In an optional embodiment, according to the first positioning data and the second positioning data, controlling the vehicle to be parked to perform the parking task until the vehicle to be parked moves into the target parking space includes:

[0027] The first positioning data and the second positioning data with the same timestamp are respectively used as the input value and the observation value of the same positioning cycle, and input into a preset Kalman filter model;

[0028] According to the output result of the Kalman filter model, the target position of each positioning cycle is obtained, wherein the target position is the position of the rear axle center of the to-be-parked vehicle relative to the ground coordinate system of the target parking space;

[0029] The vehicle to be parked is controlled to move toward the target parking space according to the target position of each positioning cycle, and the parking task is determined to be completed when the vehicle to be parked moves into the target parking space.

[0030] In a second aspect, an embodiment of the present invention further provides a vehicle, wherein parking control is performed on the vehicle by any method in the first aspect.

[0031] In a third aspect, an embodiment of the present invention further provides a parking control device, the device comprising:

[0032] A receiving module, used for receiving an execution instruction of a parking task;

[0033] An acquisition module, used for determining a target parking space for the vehicle to be parked according to the execution instruction, and acquiring first positioning data determined by the panoramic image of the vehicle to be parked based on the Ackerman steering model;

[0034] An acquisition module, used to acquire second positioning data according to the panoramic image of the vehicle, wherein the first positioning data and the second positioning data are both data of the relative position relationship between the to-be-parked vehicle and the target parking space;

[0035] The control module is used to control the vehicle to be parked to perform a parking task according to the first positioning data and the second positioning data until the vehicle to be parked moves into a target parking space.

[0036] In a fourth aspect, an embodiment of the present invention further provides an electronic device, comprising a processor and a memory, wherein the memory is coupled to the processor and stores instructions, and when the instructions are executed by the processor, the electronic device executes the steps of any one of the methods in the first aspect.

[0037] In a fifth aspect, an embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any one of the methods in the first aspect.

[0038] Compared with the prior art, the vehicle and parking control method and device of the present invention have the following advantages:

[0039] The technical solution of the present invention determines the target parking space of the vehicle to be parked according to the execution instruction after receiving the execution instruction of the parking task, obtains the vehicle panoramic image of the vehicle to be parked and the first positioning data determined based on the Ackerman steering model, and obtains the second positioning data according to the vehicle panoramic image. Since the first positioning data and the second positioning data are both data of the relative position relationship between the vehicle to be parked and the target parking space, the vehicle to be parked can be accurately controlled to perform the parking task according to the first positioning data and the second positioning data until the vehicle to be parked moves into the target parking space. The technical solution performs parking by fusing the first positioning data and the second positioning data, so that the vehicle panoramic image effectively compensates for the accumulated error of the Ackerman steering model in the middle and late stages of the parking task, improves the positioning accuracy of the vehicle during the parking process, and thus improves the accuracy of the vehicle automatic parking control. At the same time, the application of the control method does not require the installation of other sensors on existing vehicles, and large-scale automatic parking applications can be realized through software upgrades, thereby reducing the limitations of vehicle automatic parking applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the embodiments of this specification or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this specification. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0041] Figure 1 A flowchart of a parking control method provided by an embodiment of the present invention;

[0042] Figure 2 A schematic diagram of the principle of an Ackerman steering model with an offset introduced provided in an embodiment of the present invention;

[0043] Figure 3 A schematic diagram of the principle of a simplified Ackerman steering model provided by an embodiment of the present invention;

[0044] Figure 4 A schematic diagram of marking points of parking spaces with different structures provided by an embodiment of the present invention;

[0045] Figure 5 A schematic diagram of a parking space coordinate system provided by an embodiment of the present invention;

[0046] Figure 6 A position relationship diagram of vehicles and parking spaces corresponding to different heading angles provided in an embodiment of the present invention;

[0047] Figure 7 An overall logic diagram of a parking control method provided by an embodiment of the present invention;

[0048] Figure 8 A schematic structural diagram of a parking control device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0049] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention belong to the scope of protection of the embodiments of the present invention.

[0050] At present, for the automatic parking control of vehicles, some manufacturers implement pure visual solutions based on images captured by cameras. During the parking process, it is easily affected by factors such as lighting, shadows, and occlusions, resulting in the failure of tracking the target parking space mark point and the inability to achieve positioning. The following embodiments of the present invention will specifically explain how to improve the accuracy of vehicle control during the automatic parking process.

[0051] See also Figure 1 , Figure 1 A flowchart of a parking control method is provided for an embodiment of the present invention. The control method can be implemented based on a vehicle controller (ECU, Electronic Control Unit) of a vehicle, or can be implemented based on an independently configured parking controller. Any device that can run the control method is not specifically limited here. The control method includes:

[0052] S11, receiving an execution instruction of a parking task.

[0053] Specifically, the execution instruction can be obtained based on the vehicle driver's control input to the vehicle computer. When the vehicle needs to implement automatic parking, the execution instruction is sent to the parking assistance system based on the driver's gear control and the vehicle computer control. For example, the driver activates the parking assistance system by pressing a parking button or using a voice command. The parking assistance system automatically identifies a suitable target parking space and assists the driver to complete automatic parking. After receiving the execution instruction of the parking task, step S12 is entered.

[0054] S12: determining a target parking space for the vehicle to be parked according to the execution instruction, and acquiring a panoramic image of the vehicle to be parked and first positioning data determined based on an Ackerman steering model.

[0055] Specifically, the target parking space is the parking space that the vehicle to be parked needs to enter to perform the parking task, which can be determined based on the driver's operation on the vehicle's central control, or automatically determined by the parking assistance system. The vehicle panoramic image is stitched together based on images taken by multiple cameras arranged around the vehicle body, and the vehicle panoramic image can be displayed on the vehicle's central control screen.

[0056] It can be understood that the Ackerman steering model is mostly used for the front wheel steering control of four-wheel vehicles. Its core idea is to ensure that the steering angles of the inner and outer front wheels are different when the vehicle turns, so as to maintain the stability of the vehicle and reduce tire wear. The Ackerman steering model ensures that the vehicle can maintain good steering performance under various driving conditions through reasonable geometric design and electronic control. The first positioning data is obtained based on the data collected by the sensor on the vehicle and processed by the Ackerman steering model, which represents the relative position relationship between the vehicle and the target parking space.

[0057] Exemplarily, step S12 includes sub-steps S12-1 to S12-3, which are specifically described as follows:

[0058] S12-1. Processing the steering data of the vehicle to be parked based on the Ackerman steering model to obtain a first heading angle of the ground coordinate system of the vehicle to be parked at the target parking space in different positioning periods.

[0059] When implementing vehicle steering positioning based on the Ackerman steering model, the inner and outer turning angles of the vehicle's front wheels are required. This angle value cannot be directly obtained through sensors. The embodiment of the present invention obtains it based on the steering wheel angle. Specifically, it includes: obtaining the vehicle's steering wheel angle based on the vehicle's underlying CAN (Controller Area Network) message. Since there is a corresponding relationship between the steering wheel angle and the inner and outer turning angles of the front wheels when the vehicle moves forward and backward, a mapping table between the two can be obtained based on the above correspondence. The steering transmission ratio has a nonlinear characteristic. The mapping table between the steering wheel angle and the inner and outer turning angles of the front wheels is fitted using a quadratic nonlinear function. The quadratic nonlinear fitting function is:

[0060] K i =a i δ 2 +b i δ+c i

[0061] Among them, K i is the steering gear ratio, a i ,b i ,c iare the first, second and third coefficients of the quadratic nonlinear function, i is the parameter corresponding to the inner and outer turning angles of the front wheels when the vehicle is moving forward and backward, and δ is the steering wheel angle of the vehicle. The quadratic nonlinear fitting function characterizes the corresponding relationship between the steering wheel angle and the inner and outer turning angles of the front wheels, so the inner and outer turning angles of the front wheels of the vehicle can be obtained based on the steering wheel angle and the steering gear ratio.

[0062] When the Ackerman steering model implements positioning processing, it is also necessary to obtain the axle length and wheelbase of the vehicle and establish a coordinate system for data sampling. The axle length and wheelbase are input into the Ackerman steering model as calculation parameters. The coordinate system for data sampling takes the center of the rear axle of the vehicle as the coordinate origin, perpendicular to the rear axle of the vehicle, along the forward direction of the vehicle as the positive direction of the x-axis, perpendicular to the left of the vehicle body as the positive direction of the y-axis, and the initial heading angle is 90°.

[0063] In order to ensure the vehicle's maneuverability and reduce tire wear, the center of the vehicle's turning is not located on the extension line of the rear axle, but is offset y degrees backward along the longitudinal direction of the vehicle body. off See Figure 2 , Figure 2 Point O is the turning center of the vehicle, and the inner and outer turning angles of the two front wheels of the vehicle are α and β respectively. For the convenience of calculation, the embodiment of the present invention simplifies the four-wheel Ackerman steering model into a two-wheel steering model and then calculates the front wheel turning angle of the vehicle. The front wheel turning angle of the vehicle is obtained by taking the average of the inner and outer wheel turning angles of the two front wheels. Figure 3 , Figure 3 is the simplified Ackerman steering model, δ r =(α+β) / 2.

[0064] The Ackerman steering model expression at this time is:

[0065]

[0066] Among them, R r is the turning radius of the rear axle center of the vehicle with the offset introduced, y off is the offset introduced, L is the vehicle wheelbase, δ r is the front wheel turning angle.

[0067] Please continue reading Figure 3 Based on the Pythagorean theorem, the turning radius r of the vehicle front axle after the simplified Ackerman steering model can be calculated. f , the specific formula is as follows:

[0068]

[0069] It is understood that the wheels of the vehicle are in motion during the steering process, and the vehicle wheel speed can be obtained based on the pulse data collected by the wheel speed sensor installed on the wheel. Therefore, the increment of the first heading angle of the vehicle to be parked in different positioning cycles can be calculated by the following formula.

[0070]

[0071] Among them, Δθ is the increment of the first heading angle in adjacent positioning cycles, FL n and FR n are the number of pulses of the left front wheel and the right front wheel of the vehicle respectively, and Δd is the wheel travel distance of a single pulse. The first heading angle of each positioning cycle can be obtained based on the sum of the first heading angle of the previous positioning cycle and the heading angle increment from the previous positioning cycle to the current positioning cycle.

[0072] S12-2, obtaining a first coordinate value of the rear axle center of the vehicle to be parked in each positioning cycle in the ground coordinate system according to the first heading angle of each positioning cycle and the pulse data collected by the wheel speed sensor of the vehicle to be parked.

[0073] A data calculation model can be formulated based on the first heading angle, pulse data and first coordinate value. When the first coordinate value needs to be determined in each positioning cycle, the first heading angle and pulse data are input into the data calculation model, and the first coordinate value of the positioning cycle is determined through the output result of the data calculation model.

[0074] For example, according to the formula:

[0075]

[0076] Obtain the horizontal coordinate increment Δx and vertical coordinate increment Δy of the axis center in the ground coordinate system after adjacent positioning cycles, where BL n and BR n are the number of left rear wheel pulses and right rear wheel pulses of the vehicle to be parked in adjacent positioning cycles, Δd is the wheel travel distance of a single pulse, Δθ is the first heading angle increment in adjacent positioning cycles, and θ t It is understood that the calculation of the above formula is set by considering the difference in rolling distance between the front and rear wheels when the vehicle turns. Based on the above method, the horizontal coordinate increment and the vertical coordinate increment of the vehicle in the ground coordinate system can be accurately obtained.

[0077] After obtaining the abscissa increment and ordinate increment of the adjacent positioning cycle, the first coordinate value of the vehicle to be parked in each positioning cycle is obtained according to the abscissa increment, ordinate increment and the abscissa and ordinate of the vehicle to be parked in the previous positioning cycle. That is, the following relationship exists:

[0078] x t+1 =xt +Δx

[0079] y t+1 =y t +Δy

[0080] Among them, x t+1 is the current horizontal coordinate of the vehicle in the ground coordinate system in the current positioning cycle, y t+1 is the current vertical coordinate of the vehicle in the ground coordinate system in the current positioning cycle, Δx is the horizontal coordinate increment, Δy is the vertical coordinate increment, x t is the horizontal coordinate of the last positioning cycle, and Δy is the vertical coordinate of the last positioning cycle. The current horizontal coordinate and the current vertical coordinate of the current positioning cycle are the first coordinate values ​​of the vehicle to be parked in the current positioning cycle.

[0081] S12-3. Output first positioning data according to the first coordinate value.

[0082] The first coordinate value is the position of the rear axle center of the vehicle in the ground coordinate system, which is output as the first positioning data. It should be noted that the ground coordinate system is a coordinate system constructed based on the location of the target parking space. When determining the target parking space, the ground coordinate system can be established at the corner position of the target parking space based on the relative position between the vehicle and the target parking space at the determined moment, and the relative position relationship between the vehicle and the target parking space can be determined through the first positioning data.

[0083] So far, the vehicle panoramic image of the vehicle to be parked and the first positioning data determined based on the Ackerman steering model have been acquired.

[0084] S13. Obtain second positioning data according to the panoramic image of the vehicle, wherein the first positioning data and the second positioning data are both data of a relative position relationship between the vehicle to be parked and the target parking space.

[0085] Specifically, the panoramic image of the vehicle can be obtained through the panoramic surround view system, which is already widely installed on vehicles. During the parking process, the parking space markers have obvious features in the panoramic surround view image and can be well applied to detection and tracking tasks. Figure 4 , Figure 4 The parking space marking points of different types of parking spaces are shown. For L-type parking spaces, the outer left top corner can be used as the parking space marking point, and the ground coordinate system is established at this position; T-type parking spaces and oblique parking spaces also have obvious parking space marking points, which can be identified in the vehicle panoramic image through the trained convolutional neural network model.

[0086] The vehicle panoramic image can represent the relative position relationship between the vehicle and the target parking space in the form of an image. The distance and angle between the vehicle and the target parking space can be calculated based on the vehicle panoramic image, and the distance and angle are determined as the second positioning data. When obtaining the second positioning data based on the vehicle panoramic image, the vehicle panoramic image can be preprocessed first, such as performing noise reduction processing on it, to improve the image quality of the vehicle panoramic image. Then, a ground coordinate system is established based on the clearer vehicle panoramic image, and the distance and angle between the vehicle and the ground coordinate system are determined through the vehicle panoramic image, and converted into the coordinate value of the vehicle in the ground coordinate system to obtain the second positioning data of the vehicle.

[0087] Exemplarily, step S13 includes sub-steps S13-1 to S13-3, which are described in detail as follows:

[0088] S13-1. Obtain the rear axle center coordinate value of the to-be-parked vehicle in the parking space coordinate system according to the vehicle panoramic image, wherein the parking space coordinate system is a coordinate system constructed based on the location of the target parking space on the vehicle panoramic image. It can be understood that the vehicle panoramic image is obtained by dedistorting and splicing the fisheye image, so the positioning data represented by the parking space coordinate system needs to be converted to the ground coordinate system. Similarly, the parking space coordinate system can be established based on the parking space mark points of the vehicle panoramic image, and the position of the vehicle rear axle center is calculated as the vehicle position in the parking space coordinate system, and the calculation result is determined as the rear axle center coordinate value.

[0089] S13-2. Obtain a second coordinate value of the vehicle to be parked in the ground coordinate system according to the rear axle center coordinate value and a preset coordinate mapping relationship, wherein the coordinate mapping relationship is a position mapping relationship between the parking space coordinate system and the ground coordinate system.

[0090] The coordinate mapping relationship can be obtained through calibration experiments. The corresponding data is queried in the coordinate mapping relationship obtained through calibration by using the coordinate value of the rear axle center, and is used as the second coordinate value of the ground coordinate system. For example, the actual distance of the ground coordinate system is calculated by the following formula.

[0091] Δx=μ×Δx p

[0092] Among them, Δx is the actual distance of the ground coordinate system, Δx p is the pixel distance of the parking space coordinate system, and μ is the mapping relationship coefficient that characterizes the coordinate mapping relationship.

[0093] S13-3, outputting second positioning data according to the second coordinate value. When the vehicle performs the parking task, the panoramic image of the vehicle is obtained based on real-time acquisition by the camera, so the second positioning data has high accuracy, and its accuracy does not change with the execution of the parking task.

[0094] It can be understood that the ROI (Region of Interest) where the parking space mark of the target parking space is located can be determined through the panoramic image of the vehicle. When the vehicle is parked, the image frame of the detected parking space mark is tracked and predicted using a multi-target tracker, and the parking space mark detected in the current image frame is associated and matched to obtain the specific location information of the corresponding ID tracking target. Specifically, the tracker is initialized for each image frame of the detected parking space mark, a tracking ID is assigned, and the Kalman filter algorithm is used to predict the tracking target. Considering that the feature differences between the parking space mark points are small, the motion information and appearance information of the target parking space are considered at the same time during the association matching process, and the Mahalanobis distance and cosine distance between the associated features are calculated. The Hungarian algorithm is used for association matching. The above method can effectively improve the problem of ID jump of the tracking target.

[0095] When tracking the target parking space, the bounding boxes and tracking IDs of all the initialized parking space markers can be obtained. If the center point of the bounding box is within the ROI area, it is determined to be the parking space marker of the target parking space, and the tracking IDs are recorded as and And get the tracking ID of each subsequent frame in real time and The tracking box position. and The adjacent image frames having parking space marking points of the target parking space in the vehicle panoramic image are represented.

[0096] See also Figure 5 , Figure 5 The schematic diagram of the parking space coordinate system is as follows: a parking space coordinate system is established with the left mark point of the target parking space head as the origin and the parking space head entrance line as the x-axis. The second heading angle θ in the parking space coordinate system can be calculated based on the position relationship of the parking space mark points in the image. r , by rotating the coordinate system, we can get the coordinates of the center of the rear axle of the vehicle in the parking space coordinate system. The calculation formula is as follows:

[0097]

[0098] where θ r is the second heading angle of the vehicle in the parking space coordinate system of the vehicle panoramic image, X p Y is the horizontal coordinate of the center position of the rear axle of the vehicle in the ground coordinate system, p is the ordinate of the center of the rear axle of the vehicle in the ground coordinate system. p , Y p ) is recorded as the second coordinate value. Figure 6 , Figure 6is a position relationship diagram of the vehicle and parking space corresponding to different heading angles. The center position of the rear axle of the vehicle obtained in the above method is in the parking space coordinate system, and the unit is pixel. Therefore, through the coordinate mapping relationship μ and the relative position of the parking space coordinate system and the ground coordinate system at the initial moment, the coordinate origin and direction of the coordinate system are converted, and the position of the center of the rear axle of the vehicle in the vehicle ground coordinate system at the current sampling moment is calculated, that is, the second positioning data. At this point, the vehicle positioning process based on the tracking of parking space markers on the vehicle panoramic image is completed.

[0099] S14. Control the vehicle to be parked to perform a parking task according to the first positioning data and the second positioning data until the vehicle to be parked moves into a target parking space.

[0100] Specifically, since the Ackerman steering model is based on the ground coordinate system constructed at the time of determining the target parking space, there is a cumulative error as the parking task proceeds, and the dynamic movement of the vehicle will cause the first positioning data determined by the Ackerman steering model to have insufficient accuracy. The vehicle positioning method based on the tracking of parking space markers on the vehicle panoramic image is a visual method. When the parking space markers of the target parking space exceed the field of view of the image, it will cause the tracking to be lost and the vehicle position cannot be output. Therefore, it is necessary to correct the second positioning data of the vehicle panoramic image by fusing the first positioning data. Different weights can be configured based on the execution node of the parking task. For example, in the initial stage of the parking task, the configuration weight of the first positioning data is greater than the configuration weight of the second positioning data; in the middle stage of the parking task, the configuration weights of the first positioning data and the second positioning data are the same; in the final stage of the parking task, the configuration weight of the first positioning data is less than the configuration weight of the second positioning data. Give full play to the advantages of different positioning data until the vehicle to be parked moves to the target parking space.

[0101] In actual applications, the first positioning data is sent via CAN messages, while the output response time of the second positioning data is slow. If the fusion of the first positioning data and the second positioning data is directly implemented, there will be confusion. Based on this, before implementing data fusion, the first positioning data with the same timestamp is determined in the data queue of the first positioning data according to the timestamp of the second positioning data; the first positioning data and the second positioning data with the same timestamp are determined as the target data for the parking task of the vehicle to be parked. By determining the first positioning data and the second positioning data at the same time through the timestamp, the positioning data can be synchronized, providing accuracy and robustness during the execution of the parking task.

[0102] Exemplarily, step S14 includes sub-steps S14-1 to S14-3, which are specifically described as follows: including:

[0103] S14-1. The first positioning data and the second positioning data with the same timestamp are used as the input value and the observation value of the same positioning cycle, respectively, and input into the preset Kalman filter model. Kalman filtering is an efficient recursive filter that can estimate the state of a dynamic system in a noisy environment. Its core function is to combine the dynamic model of the system and the observation data to estimate the true state of the system mathematically. The Kalman filter estimates the state of the next moment through the prediction step, and then uses the actual observation value to correct the prediction error through the update step, so as to obtain a more accurate estimate. This process is continuously iterated, so that the filter can adapt to the changes in the system state and the uncertainty of the observation noise.

[0104] S14-2. According to the output result of the Kalman filter model, the target position of each positioning cycle is obtained, wherein the target position is the position of the rear axle center of the to-be-parked vehicle relative to the ground coordinate system of the target parking space. The Kalman filter model predicts the dynamic changes of the parking task by iterative prediction and update steps, using the observed value and the control input value, estimates the next state through the prediction step, and corrects the prediction by the observed value in the update step, and finally outputs the target value. The position and heading angle changes of the vehicle during the parking process are nonlinear. The embodiment of the present invention uses the extended Kalman filter method for data fusion, performs state prediction on the vehicle positioning information based on the Ackerman steering model, and uses the vehicle positioning information based on the parking space mark point tracking as the real-time observation. Finally, the optimal position estimate at the current moment can be updated by combining the predicted value with the observed value at the current moment to accurately obtain the target position of each positioning cycle.

[0105] S14-3, control the vehicle to be parked to move to the target parking space according to the target position of each positioning cycle, and determine that the parking task is completed when the vehicle to be parked moves into the target parking space. It can be understood that the fusion of the positioning data of the two positioning methods effectively solves the problem of cumulative error in the vehicle positioning method based on the Ackerman steering model and the problem of tracking failure and failure to cover the entire parking process in the vehicle positioning method based on tracking of parking space markers on the panoramic surround image, thereby improving the positioning accuracy of the vehicle during the parking process.

[0106] The following embodiment of the present invention will generally explain how to integrate two different positioning data to perform the parking task of the vehicle. Figure 7 , Figure 7 This is the overall logic diagram of the parking control method. Specifically, it includes:

[0107] S701: Construct a first positioning module based on an Ackerman steering model.

[0108] S702: Obtain a mapping relationship between a steering wheel angle of the vehicle and an inner and outer corner angle of a front wheel.

[0109] S703: Obtain Ackerman steering model parameters and establish a sampling coordinate system.

[0110] S704: Calculate the increment of the vehicle heading angle using the front wheel pulses.

[0111] S705 . Calculate the increments of the center of the rear axle of the vehicle in the x-direction and the y-direction using the rear wheel pulses.

[0112] S706: Calculate and obtain the center position of the rear axle of the vehicle at the current sampling time, that is, the first positioning data, based on the above sampling time positions.

[0113] S707: Construct a second positioning module based on tracking parking space marker points on the vehicle panoramic image.

[0114] S708: Obtain a mapping relationship between a pixel distance on the vehicle panoramic image and an actual distance on the real ground.

[0115] S709: Use a multi-target tracker to simultaneously detect parking space marking points on the vehicle panoramic image.

[0116] S710: Use a multi-target tracker to track and predict the detected parking space marking points to obtain the position of the target ID parking space marking points.

[0117] S711. Calculate the vehicle heading angle and the coordinates of the rear axle center in the parking space coordinate system.

[0118] S712. Obtain the position of the center of the rear axle of the vehicle in the ground coordinate system at the current time through coordinate system conversion, that is, the second positioning data.

[0119] S713: Process the first positioning data and the second positioning data by using an extended Kalman filter.

[0120] S714. Based on the output result of the extended Kalman filter, the fused real-time position of the vehicle is obtained, and the vehicle parking task is executed until the vehicle to be parked moves to the target parking space.

[0121] Based on the same technical concept as the control method, an embodiment of the present invention further provides a vehicle, which is parked under control of any method.

[0122] Based on the same technical concept as the control method, the embodiment of the present invention also provides a parking control device, see Figure 8 , Figure 8 The schematic diagram of the parking control device is shown in FIG. 1 , wherein the control device comprises:

[0123] A receiving module 801 is used to receive an execution instruction of a parking task;

[0124] An acquisition module 802 is used to determine a target parking space for the vehicle to be parked according to the execution instruction, and to acquire first positioning data of the panoramic image of the vehicle to be parked determined based on the Ackerman steering model;

[0125] The obtaining module 803 is used to obtain second positioning data according to the panoramic image of the vehicle, wherein the first positioning data and the second positioning data are both data of the relative position relationship between the to-be-parked vehicle and the target parking space;

[0126] The control module 804 is used to control the vehicle to be parked to perform the parking task according to the first positioning data and the second positioning data until the vehicle to be parked moves to the target parking space.

[0127] In an optional embodiment, the acquisition module includes:

[0128] A first obtaining submodule is used to process the steering data of the vehicle to be parked based on the Ackerman steering model to obtain a first heading angle of the ground coordinate system of the vehicle to be parked at the target parking space in different positioning periods;

[0129] The second obtaining submodule is used to obtain the first coordinate value of the rear axle center of the vehicle to be parked in each positioning period in the ground coordinate system according to the first heading angle of each positioning period and the pulse data collected by the wheel speed sensor on the vehicle to be parked;

[0130] The first output submodule is used to output first positioning data according to the first coordinate value.

[0131] In an optional embodiment, the second obtaining submodule includes:

[0132] The first unit is used according to the formula:

[0133]

[0134] Obtain the horizontal coordinate increment Δx and vertical coordinate increment Δy of the axis center in the ground coordinate system after adjacent positioning cycles, where BL n and BR n are the number of left rear wheel pulses and right rear wheel pulses of the vehicle to be parked in adjacent positioning cycles, Δd is the wheel travel distance of a single pulse, Δθ is the first heading angle increment in adjacent positioning cycles, and θ t is the first heading angle of the previous positioning cycle;

[0135] The second obtaining unit is used to obtain the first coordinate value of the vehicle to be parked in each positioning cycle according to the abscissa increment and ordinate increment of adjacent positioning cycles and the abscissa and ordinate of the vehicle to be parked in the previous positioning cycle.

[0136] In an optional embodiment, the obtaining module includes:

[0137] A third obtaining submodule is used to obtain the rear axle center coordinate value of the to-be-parked vehicle in the parking space coordinate system according to the vehicle panoramic image, wherein the parking space coordinate system is a coordinate system constructed based on the position of the target parking space on the vehicle panoramic image;

[0138] A fourth obtaining submodule is used to obtain a second coordinate value of the vehicle to be parked in the ground coordinate system according to the rear axle center coordinate value and a preset coordinate mapping relationship, wherein the coordinate mapping relationship is a position mapping relationship between the parking space coordinate system and the ground coordinate system;

[0139] The second output submodule is used to output second positioning data according to the second coordinate value.

[0140] In an optional embodiment, the control device further includes:

[0141] A first determining module, configured to determine, in a data queue of the first positioning data, first positioning data having the same timestamp as the timestamp of the second positioning data;

[0142] The second determining module is used to determine the first positioning data and the second positioning data with the same timestamp as target data for executing the parking task for the vehicle to be parked.

[0143] In an optional embodiment, the control module includes:

[0144] An input submodule, used to input the first positioning data and the second positioning data with the same timestamp as the input value and the observation value of the same positioning cycle respectively, into a preset Kalman filter model;

[0145] A fifth obtaining submodule is used to obtain a target position of each positioning cycle according to an output result of the Kalman filter model, wherein the target position is a position of a rear axle center of the to-be-parked vehicle relative to a ground coordinate system of a target parking space;

[0146] The determination submodule is used to control the vehicle to be parked to move toward the target parking space according to the target position of each positioning cycle, and determine that the parking task is completed when the vehicle to be parked moves into the target parking space.

[0147] Based on the same technical concept as the control method, an embodiment of the present invention also provides an electronic device, including a processor and a memory, wherein the memory is coupled to the processor and stores instructions, and when the instructions are executed by the processor, the electronic device executes the steps of any one of the control methods.

[0148] Based on the same technical concept as the control method, an embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the steps of any one of the control methods are implemented.

[0149] The technical solution provided in the embodiments of the present invention has at least the following technical effects or advantages:

[0150] After receiving the execution instruction of the parking task, the target parking space of the vehicle to be parked is determined according to the execution instruction, and the vehicle panoramic image of the vehicle to be parked and the first positioning data determined based on the Ackerman steering model are obtained. The second positioning data is obtained according to the vehicle panoramic image. Since the first positioning data and the second positioning data are both data of the relative position relationship between the vehicle to be parked and the target parking space, the vehicle to be parked can be accurately controlled to perform the parking task according to the first positioning data and the second positioning data until the vehicle to be parked moves into the target parking space. This technical solution performs parking by fusing the first positioning data and the second positioning data, so that the vehicle panoramic image effectively compensates for the accumulated error of the Ackerman steering model in the middle and late stages of the parking task, improves the positioning accuracy of the vehicle during the parking process, and thus improves the accuracy of the vehicle automatic parking control. At the same time, the application of this control method does not require the installation of other sensors on existing vehicles, and large-scale automatic parking applications can be realized through software upgrades, thereby reducing the limitations of vehicle automatic parking applications.

[0151] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0152] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (modules, systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0153] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0154] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0155] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0156] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A parking control method, characterized in that: The method comprises: Receiving execution instructions of parking tasks; Determining a target parking space for the vehicle to be parked according to the execution instruction, and acquiring a panoramic image of the vehicle to be parked and first positioning data determined based on an Ackerman steering model; Obtaining second positioning data according to the vehicle panoramic image, wherein the first positioning data and the second positioning data are both data of the relative position relationship between the to-be-parked vehicle and the target parking space; According to the first positioning data and the second positioning data, the vehicle to be parked is controlled to perform the parking task until the vehicle to be parked moves into the target parking space.

2. The parking control method according to claim 1, characterized in that: Acquiring first positioning data of the vehicle to be parked determined based on the Ackerman steering model, including: Processing the steering data of the vehicle to be parked based on the Ackerman steering model to obtain a first heading angle of the vehicle to be parked in the ground coordinate system of the target parking space at different positioning periods; Obtaining a first coordinate value of the rear axle center of the vehicle to be parked in each positioning cycle in the ground coordinate system according to the first heading angle of each positioning cycle and the pulse data collected by the wheel speed sensor of the vehicle to be parked; The first positioning data is output according to the first coordinate value.

3. The parking control method according to claim 2, characterized in that: The step of obtaining a first coordinate value of a rear axle center of the vehicle to be parked in each positioning cycle in the ground coordinate system according to the first heading angle of each positioning cycle and the pulse data collected by the wheel speed sensor of the vehicle to be parked comprises: According to the formula: Obtain the horizontal coordinate increment Δx and the vertical coordinate increment Δy of the rear axle center in the ground coordinate system in adjacent positioning cycles, where BL n and BR n are the number of left rear wheel pulses and the number of right rear wheel pulses of the vehicle to be parked in adjacent positioning cycles, Δd is the wheel travel distance of a single pulse, Δθ is the first heading angle increment of the adjacent positioning cycles, θ t is the first heading angle of the previous positioning cycle; The first coordinate value of the vehicle to be parked in each positioning cycle is obtained according to the abscissa increments and ordinate increments of the adjacent positioning cycles and the abscissa and ordinate of the vehicle to be parked in the previous positioning cycle.

4. The parking control method according to claim 1, characterized in that: The obtaining of second positioning data according to the vehicle panoramic image comprises: Obtaining a rear axle center coordinate value of the to-be-parked vehicle in a parking space coordinate system according to the vehicle panoramic image, wherein the parking space coordinate system is a coordinate system constructed based on the location of the target parking space on the vehicle panoramic image; According to the rear axle center coordinate value and a preset coordinate mapping relationship, a second coordinate value of the vehicle to be parked in the ground coordinate system is obtained, wherein the coordinate mapping relationship is a position mapping relationship between the parking space coordinate system and the ground coordinate system; The second positioning data is output according to the second coordinate value.

5. The parking control method according to claim 1, characterized in that: Before the vehicle to be parked is controlled to perform the parking task according to the first positioning data and the second positioning data, the method further comprises: According to the timestamp of the second positioning data, determining the first positioning data having the same timestamp in the data queue of the first positioning data; The first positioning data and the second positioning data with the same timestamp are determined as target data for the vehicle to be parked to perform the parking task.

6. The parking control method according to claim 1, characterized in that: The controlling the vehicle to be parked to perform the parking task according to the first positioning data and the second positioning data until the vehicle to be parked moves into the target parking space includes: The first positioning data and the second positioning data with the same timestamp are respectively used as the input value and the observation value of the same positioning cycle, and input into a preset Kalman filter model; According to the output result of the Kalman filter model, a target position of each positioning cycle is obtained, wherein the target position is the position of the rear axle center of the vehicle to be parked relative to the ground coordinate system of the target parking space; The vehicle to be parked is controlled to move toward the target parking space according to the target position of each positioning cycle, and the parking task is determined to be completed when the vehicle to be parked moves into the target parking space.

7. A vehicle, characterized in that: The vehicle is parked by the method described in any one of claims 1-6.

8. A parking control device, characterized in that: The device comprises: A receiving module, used for receiving an execution instruction of a parking task; an acquisition module, configured to determine a target parking space for the vehicle to be parked according to the execution instruction, and to acquire first positioning data of the panoramic image of the vehicle to be parked determined based on an Ackerman steering model; an acquisition module, configured to acquire second positioning data according to the vehicle panoramic image, wherein the first positioning data and the second positioning data are both data on the relative position relationship between the to-be-parked vehicle and the target parking space; A control module is used to control the vehicle to be parked to perform the parking task according to the first positioning data and the second positioning data until the vehicle to be parked moves into the target parking space.

9. An electronic device, characterized in that: The electronic device comprises a processor and a memory, wherein the memory is coupled to the processor and stores instructions, and when the instructions are executed by the processor, the electronic device executes the steps of the method according to any one of claims 1 to 6.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method described in any one of claims 1 to 6 are implemented.