Automatic parking control method, electronic device, storage medium and program product

Abstract

Embodiments of the present application provide an automatic parking control method, electronic device, storage medium, and program product. The method includes: in response to an automatic parking command, identifying a parking area based on sensor data, and performing path planning based on the parking area to obtain a reference path; during automatic parking based on the reference path, determining the heading deviation, lateral deviation, longitudinal deviation, and speed deviation based on the vehicle's real-time data and the reference path; processing the heading deviation using a linear control strategy to obtain a heading steering angle command; processing the heading deviation, lateral deviation, and longitudinal deviation using a nonlinear control strategy to obtain a lateral steering angle command; determining a steering angle command based on the heading steering angle command and the lateral steering angle command; processing the speed deviation using a linear control strategy to obtain a speed command; and adjusting the vehicle's steering angle and speed based on the steering angle command and the speed command. This method is used to achieve precise path tracking.

Description

Technical Field

[0001] The present invention relates to the field of intelligent driving control technology, and in particular to an automatic parking control method, electronic equipment, storage medium, and program product. Background Art

[0002] With the development of technology, automatic parking function has become an indispensable intelligent driving function.

[0003] In related technologies, a parking path is generated through a path planning algorithm, and the vehicle is controlled to perform automatic parking according to the parking path. However, in the process of controlling the vehicle to perform automatic parking according to the parking path, it is easy for the actual driving path of the vehicle to deviate from the planned path, making it difficult to park according to the planned path. Summary of the Invention

[0004] The embodiments of the present application provide an automatic parking control method, an electronic device, a storage medium, and a program product for quickly and stably correcting deviations during the automatic parking process to achieve accurate path tracking.

[0005] In a first aspect, an embodiment of the present application provides an automatic parking control method, comprising: in response to an automatic parking instruction, identifying a parking area based on sensor data, and performing path planning based on the parking area to obtain a reference path; in the process of automatic parking based on the reference path, determining a deviation set according to the vehicle's real-time data and the reference path; the deviation set includes a heading deviation, a lateral deviation, a longitudinal deviation, and a speed deviation; processing the heading deviation using a linear control strategy to obtain a heading steering angle instruction; fusing the heading deviation, the lateral deviation, and the longitudinal deviation using a nonlinear control strategy to obtain a lateral steering angle instruction; determining a steering angle instruction according to the heading steering angle instruction and the lateral steering angle instruction; processing the speed deviation according to the linear control strategy to obtain a speed instruction; adjusting the steering angle and vehicle speed of the vehicle based on the steering angle instruction and the speed instruction; and stopping the automatic parking when it is determined that the parking completion condition is met based on the target distance between the vehicle and the parking area.

[0006] In a second aspect, an embodiment of the present application provides an automatic parking control device, comprising:

[0007] a path planning module, configured to identify a parking area based on sensor data in response to an automatic parking instruction, and perform path planning based on the parking area to obtain a reference path;

[0008] a deviation determination module for determining a deviation set based on the vehicle's real-time data and the reference path during automatic parking based on the reference path; the deviation set includes heading deviation, lateral deviation, longitudinal deviation, and speed deviation;

[0009] The command determination module is used to process the heading deviation using a linear control strategy to obtain a heading steering angle command; to fuse the heading deviation, lateral deviation, and longitudinal deviation using a nonlinear control strategy to obtain a lateral steering angle command; to determine the steering angle command based on the heading steering angle command and the lateral steering angle command; and to process the speed deviation using a linear control strategy to obtain a speed command.

[0010] A dynamic adjustment module, for adjusting the steering angle and speed of the vehicle based on the steering angle command and the speed command;

[0011] The parking stop module is configured to stop the automatic parking when it is determined that a parking completion condition is satisfied based on a target distance between the vehicle and the parking area.

[0012] In a third aspect, an embodiment of the present application provides an electronic device comprising: a memory, a processor; the memory stores computer-executable instructions; the processor executes the computer-executable instructions stored in the memory, so that the processor executes the first aspect above and / or various possible implementations of the first aspect.

[0013] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, they are used to implement the first aspect above and / or various possible implementation methods of the first aspect.

[0014] In a fifth aspect, an embodiment of the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the above first aspect and / or various possible implementation methods of the first aspect.

[0015] Beneficial effects of the present invention:

[0016] The automatic parking control method proposed in the embodiment of the present application responds to the automatic parking instruction, identifies the parking area based on the sensor data, and performs path planning based on the parking area to obtain a reference path. In the process of automatic parking based on the reference path, a deviation set is determined according to the real-time data of the vehicle and the reference path, and the deviation set includes a heading deviation, a lateral deviation, a longitudinal deviation and a speed deviation; a linear control strategy is used to process the heading deviation to obtain a heading steering angle instruction; a nonlinear control strategy is used to fuse the heading deviation, the lateral deviation and the longitudinal deviation to obtain a lateral steering angle instruction; and a heading steering angle instruction is obtained according to the heading steering angle instruction and the lateral steering angle instruction. The steering angle command determines the steering angle command; the speed deviation is processed according to the linear control strategy to obtain the speed command, and the steering angle and speed of the vehicle are adjusted based on the steering angle command and the speed command; in the parking scenario, the nonlinear control strategy enables the automatic parking to have a strong path following capability, and the linear control strategy can respond quickly and eliminate steady-state errors. In the process of parking according to the reference path, the embodiment of the present application combines the nonlinear control strategy with the linear control strategy, which can quickly and stably correct the deviation, accurately achieve path tracking, and provide smooth steering control, thereby improving the accuracy and robustness of automatic parking. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 Schematic diagram of the process of the automatic parking control method provided by the present invention Figure 1 ;

[0018] Figure 2 A schematic diagram of a process for generating a reference path provided by the present invention;

[0019] Figure 3 A schematic diagram of determining lateral deviation, longitudinal deviation and heading deviation provided by the present invention;

[0020] Figure 4 Schematic diagram of the process of the automatic parking control method provided by the present invention Figure 2 ;

[0021] Figure 5 A schematic diagram of the speed adjustment area provided by the present invention;

[0022] Figure 6 A schematic diagram of a ground scene without parking space lines provided by the present invention;

[0023] Figure 7 A schematic diagram of the process of identifying a parking area provided by the present invention;

[0024] Figure 8 A schematic structural diagram of the automatic parking control device provided by the present invention;

[0025] Figure 9 This is a schematic diagram of the structure of the electronic device provided in this application. DETAILED DESCRIPTION

[0026] Exemplary embodiments will be described in detail herein, with examples illustrated in the accompanying drawings. In the following description, when referring to the drawings, identical numerals in different figures represent identical or similar elements, unless otherwise indicated. The embodiments described in the following exemplary embodiments are not intended to represent all embodiments consistent with the present application. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present application, as detailed in the appended claims.

[0027] The following specific embodiments describe in detail the technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0028] Figure 1 Schematic diagram of the process of the automatic parking control method provided in this application Figure 1 , the automatic parking control method can be applied to electronic equipment, and the electronic equipment can be a vehicle end; Figure 1 As shown, the automatic parking control method includes:

[0029] S101 . In response to an automatic parking instruction, identify a parking area based on sensor data, and perform path planning based on the parking area to obtain a reference path.

[0030] Among them, the automatic parking instruction is used to instruct the start of automatic parking; automatic parking includes: detecting the parking area, planning the parking path, and controlling the vehicle to park in the parking area according to the parking path.

[0031] The sensor data is obtained by on-board sensors; on-board sensors include but are not limited to: multiple on-board cameras with different perspectives, ultrasonic radars and lidars; accordingly, the sensor data includes: images obtained by on-board cameras, distance data obtained by ultrasonic radars, and point cloud data obtained by lidars.

[0032] The parking area is an area used for parking; the reference path is the driving path for parking the vehicle from the current position into the parking area; the reference path includes: multiple reference path points forming the reference path, a reference heading angle, a reference vehicle speed and a reference curvature for each reference path point; the multiple reference path points include a reference starting point, a reference end point and other reference path points; the reference starting point corresponds to the current position of the vehicle, and the reference end point corresponds to the position within the parking area; other reference path points are reference path points other than the reference starting point and the reference end point.

[0033] Optionally, in a parking scenario where the vehicle is in a parking space with parking lines, ground parking lines and obstacles are detected based on sensor data. When at least two parallel or vertical parking lines are detected, there are no obstacles between the parking lines, and the spacing between the parking lines meets a preset threshold, the area between the detected parking lines is marked as a candidate parking space area; wherein the preset threshold can be set according to actual needs, and the embodiment of the present application does not limit this; the parking area is determined in the candidate parking space area based on the width and length of the candidate parking space area.

[0034] Optionally, in a parking scenario where the vehicle is in a parking area without parking space lines, target detection is performed based on sensor data to determine a blank area that does not include obstacles, and a candidate parking area whose size meets the parking conditions is selected from the blank area, and a parking area is selected from the candidate parking area based on the flatness of the candidate parking area; wherein, the candidate parking area whose size meets the parking conditions refers to the candidate parking area with the minimum length in the first preset direction, which is greater than the length threshold, and the candidate parking area with the minimum width in the second preset direction, which is greater than the width threshold; the length threshold and the width threshold can be set according to actual needs; the first preset direction and the second preset direction are orthogonal to each other, and the specific directions of the first preset direction and the second preset direction can be set according to actual needs, which is not limited in the embodiment of the present application; it should be noted that the minimum length of the candidate parking area in the first preset direction is greater than the length threshold, and the minimum width of the candidate parking area in the second preset direction is greater than the width threshold, indicating that the maximum inscribed rectangle of the candidate parking area can accommodate parking vehicles.

[0035] After the parking area is determined, path planning is performed based on the parking area, the current position of the vehicle, and obstacles to obtain a reference path.

[0036] Specifically, if Figure 2 As shown in the figure, a grid map is constructed based on the parking area and obstacle information, and an initial global path is generated using the path planning algorithm and the grid map. The path segments in the initial global path are locally optimized to obtain an initial smooth curve. The velocity profile of each point in the initial smooth curve is then determined to obtain a reference path. Local optimization of the path segments in the initial global path can adjust the smoothness of the initial path and reduce sharp turns and redundant movements in the initial path.

[0037] For example, the coordinates of the grid map are determined with the vehicle's forward direction as the +X axis and the left side as the +Y axis. The grid map is constructed based on the parking area and obstacle information. The area corresponding to the grid map can be 10m×10m and the resolution can be 0.1m.

[0038] In the grid map, grids with obstacles are marked as obstacles (1), static obstacles are marked with static symbols on the grids, dynamic obstacles are marked with dynamic symbols on the grids, and the temporarily occupied grids corresponding to the predicted trajectories of dynamic obstacles are marked; grids without obstacles are marked as idle (0), and grids where it is uncertain whether there are obstacles are marked as unknown (-1).

[0039] Identify the idle area set based on the grid map, and determine the length, width, ground flatness, and minimum distance to obstacles of each idle area set; determine the first score of the idle area set based on whether the length and width of the idle area set are greater than the sum of the vehicle body diagonal and the safety margin. For example, if the length and width of the idle area set are both greater than the sum of the vehicle body diagonal and the safety margin, the first score is the first preset score. If at least one of the length and width of the idle area set is not greater than the sum of the vehicle body diagonal and the safety margin, the first score is 0; determine the second score based on whether the ground flatness of the idle area set is less than the safety flatness. For example, if the length and width of the idle area set are both greater than the sum of the vehicle body diagonal and the safety margin, the first score is 0. If the ground flatness is less than the safety flatness, the second score is the second preset score. If the ground flatness of the free area set is not less than the safety flatness, the second score is 0. The third score is determined based on whether the minimum distance between the free area set and the obstacle is greater than the safety distance. For example, if the minimum distance between the free area set and the obstacle is greater than the safety distance, the third score is the third preset score. If the minimum distance between the free area set and the obstacle is not greater than the safety distance, the third score is 0. The first preset score, the second preset score, and the third preset score can be set according to actual needs. For example, the first preset score, the second preset score, and the third preset score can all be 1.

[0040] Based on the first, second, and third scores, a comprehensive score of the idle area set is determined. For example, the sum of the first, second, and third scores is used as the comprehensive score. The idle area set with the highest comprehensive score is used as the target idle area set, and the centroid of the target idle area set is used as the global target point.

[0041] The global target point is used to construct the heuristic cost function, and the A-algorithm is used to search the global path. In the process of using the A-algorithm to search the global path, the heuristic cost function and the driving cost function are used for evaluation to obtain the initial global path. Among them, the A-algorithm (A-Star Algorithm) is also called A The search algorithm is a classic path planning algorithm.

[0042] Bezier curve smoothing: After obtaining the initial global path, the local path is optimized and smoothed. Specifically, a quadratic Bezier curve segment is generated for every three adjacent points in the initial global path. All quadratic Bezier curve segments are connected in series to obtain the initial smooth curve, which can eliminate the broken line corners in the initial global path.

[0043] Perform curvature constraint: For each quadratic Bezier curve segment in the initial smooth curve, calculate the maximum curvature of the quadratic Bezier curve segment. If the maximum curvature is greater than the maximum allowable curvature of the vehicle, interpolate the quadratic Bezier curve segment. For example, you can reinsert intermediate control points in the segment or use cubic Bezier spline interpolation to smooth it again to obtain the target smooth curve.

[0044] Generate a speed profile: For each point in the target smooth curve, determine the safe speed based on the curvature of the point; add deceleration segments at the start and end points of each curve segment in the target smooth curve to ensure that deceleration is completed before entering the high curvature segment.

[0045] After Bezier curve smoothing, curvature constraint and velocity profile generation, the reference path is obtained.

[0046] S102. During automatic parking based on a reference path, determine a deviation set based on real-time vehicle data and the reference path; the deviation set includes a heading deviation, a lateral deviation, a longitudinal deviation, and a speed deviation.

[0047] Among them, the vehicle real-time data includes real-time speed, real-time position and real-time heading angle.

[0048] Specifically, during the process of automatic parking based on a reference path, the real-time vehicle speed, real-time position and real-time heading angle are obtained. The real-time position of the vehicle can be obtained through the vehicle's real-time positioning module such as fused GPS, inertial measurement unit (IMU), visual SLAM, etc.

[0049] According to the real-time position of the vehicle, a projection point on the reference path is determined; the reference path contains multiple reference path points, and the projection point is the reference path point closest to the real-time position of the vehicle among the multiple reference path points; the real-time vehicle speed is obtained, and the difference between the real-time vehicle speed and the reference vehicle speed of the projection point is determined to obtain the speed deviation; the real-time position of the vehicle is obtained, and the distance between the real-time position of the vehicle and the projection point is determined to obtain the lateral deviation; along the direction of the reference path, the distance from the real-time position of the vehicle to the end point of the reference path is determined to obtain the longitudinal deviation; the real-time heading angle is obtained, and the angle difference between the real-time heading angle of the vehicle and the reference heading angle of the projection point is determined to obtain the heading deviation.

[0050] For example, Figure 3 As shown, the projection point of the vehicle's real-time position p on the reference path is d1; the lateral deviation , is the distance between the vehicle's real-time position p and the projection point d1; longitudinal deviation , is the distance between the vehicle's real-time position p and the reference path end point dn, and the heading deviation , is the angular difference between the vehicle's real-time heading angle and the reference heading angle of the projection point d1.

[0051] In actual applications, the lateral deviation, longitudinal deviation, and heading deviation have allowable error ranges. For example, the error range of the lateral deviation is -10cm to 10cm (the error range is ≤±10cm); the error range of the longitudinal deviation is -15cm to 15cm (the error range is ≤±15cm); and the error range of the heading deviation is -5° to 5° (the error range is ≤5°).

[0052] S103. Process the heading deviation using a linear control strategy to obtain a heading steering angle command; use a nonlinear control strategy to fuse the heading deviation, lateral deviation, and longitudinal deviation to obtain a lateral steering angle command; determine a steering angle command based on the heading steering angle command and the lateral steering angle command; and process the speed deviation using a linear control strategy to obtain a speed command.

[0053] Among them, the linear control strategy means that the output of the linear control strategy is determined according to the linear combination of the deviations included in the deviation set; the nonlinear control strategy means that the output of the nonlinear control strategy is determined according to the nonlinear fusion of the deviations.

[0054] For example, the nonlinear control strategy may be a Stanley control strategy, and the linear control strategy may be a proportional-integral-derivative control strategy (PID).

[0055] Optionally, a PID controller is used to process the heading deviation to obtain a heading steering angle instruction; a Stanley controller is used to process the lateral deviation, longitudinal deviation and heading deviation to obtain a lateral steering angle instruction, and the heading steering angle instruction and the lateral steering angle instruction are weightedly summed using the preset weights of the PID controller and the preset weights of the Stanley controller to obtain a steering angle instruction; the speed deviation is processed according to the PID controller to obtain a speed instruction.

[0056] Specifically, the heading deviation is input into the heading PID controller to obtain a heading steering angle instruction, and the speed deviation is input into the speed PID controller to obtain a speed instruction.

[0057] Optionally, the lateral deviation, longitudinal deviation and heading deviation are input into the Stanley controller to obtain a lateral steering angle command.

[0058] Optionally, a target gain is determined according to the longitudinal deviation and a preset gain, a target heading deviation is determined according to the target gain and the heading deviation, and the target heading deviation and the lateral deviation are input into the Stanley controller to obtain a lateral steering angle command.

[0059] Optionally, a target gain is determined according to the longitudinal deviation and a preset gain, a target lateral deviation is determined according to the target gain and the lateral deviation, and the heading deviation and the target lateral deviation are input into a Stanley controller to obtain a lateral steering angle command.

[0060] The steering angle command is determined based on the heading steering angle command and the lateral steering angle command. This can be done by adding the heading steering angle command and the lateral steering angle command to obtain the steering angle command. Alternatively, the steering angle command can be obtained by weighted summing the heading steering angle command and the lateral steering angle command using the preset weights of the PID controller and the preset weights of the Stanley controller.

[0061] Optionally, the preset weights of the PID controller and the preset weights of the Stanley controller may be fixed values ​​set according to actual needs.

[0062] Optionally, when the longitudinal deviation falls within the first interval (indicating a vehicle is farther from the parking area), the preset weight of the PID controller is determined to be the first value, and the preset weight of the Stanley controller is determined to be the second value, where the first value is greater than the second value. When the longitudinal deviation falls within the second interval (indicating a vehicle is closer to the parking area), the preset weight of the PID controller is determined to be the second value, and the preset weight of the Stanley controller is determined to be the first value. In other words, the preset weights of the PID controller and the Stanley controller are dynamically adjusted based on the distance between the vehicle and the parking area. The first interval, the second interval, the first value, and the second value can all be set according to actual needs and are not limited in this embodiment of the present application.

[0063] Optionally, the vehicle speed, lateral deviation and heading deviation are processed according to fuzzy rules to obtain the preset weights of the PID controller and the preset weights of the Stanley controller. The fuzzy rules can be: when the vehicle speed is low and the lateral deviation is large, increase the preset weight of the Stanley controller; when the vehicle speed is high and the lateral deviation is small, increase the preset weight of the PID controller.

[0064] Among them, a low vehicle speed and a high vehicle speed can be measured by a preset vehicle speed range. For example, when the vehicle speed belongs to the preset vehicle speed range, the vehicle speed is determined to be low, and when the vehicle speed does not belong to the preset vehicle speed range, the vehicle speed is determined to be high; the preset vehicle speed range can be set according to actual needs, and the embodiment of the present application does not limit the specific value of the preset vehicle speed range; a large lateral deviation and a small lateral deviation can be measured by a preset lateral deviation range. For example, when the lateral deviation belongs to the preset lateral deviation range, the lateral deviation is determined to be small, and when the lateral deviation does not belong to the preset lateral deviation, the lateral deviation is determined to be large; the preset lateral deviation range can be set according to actual needs, and the embodiment of the present application does not limit the specific value of the preset lateral deviation range.

[0065] Increasing the preset weight of the Stanley controller may be to increase the fuzzy weight increment for the preset weight of the Stanley controller, and increasing the preset weight of the PID controller may be to increase the fuzzy weight increment for the preset weight of the PID controller; the fuzzy weight increment may be determined by determining a preset fuzzy factor corresponding to the vehicle speed, calculating the product between the preset fuzzy factor and the preset weight increment, and using the product as the fuzzy weight increment; the preset fuzzy factor and the preset weight increment may be set according to actual needs, and the embodiment of the present application does not limit the specific values ​​of the preset fuzzy factor and the preset weight increment.

[0066] Optionally, a PID controller is used to process the lateral deviation to obtain a first steering angle instruction; a Stanley controller is used to process the lateral deviation and the heading deviation to obtain a second steering angle instruction; the first steering angle instruction and the second steering angle instruction are weightedly summed using the preset weights of the PID controller and the preset weights of the Stanley controller to obtain a steering angle instruction; and the speed deviation is processed according to the PID controller to obtain a speed instruction.

[0067] S104: Adjust the steering angle and vehicle speed of the vehicle based on the steering angle command and the speed command.

[0068] Among them, the steering angle instruction includes: front wheel steering angle and steering direction; the front wheel steering angle is represented by a specific angle value, and the steering direction can be: turn left, turn right or go straight; the speed instruction includes: target speed, target acceleration and vehicle speed direction; the target speed represents the speed that the vehicle needs to reach, and the target acceleration represents the acceleration required to reach the target speed; the target acceleration can be a positive value (indicating that the vehicle needs to accelerate) or a negative value (indicating that the vehicle needs to decelerate); the vehicle speed direction is used to indicate whether the vehicle is moving forward or reverse.

[0069] Specifically, after determining the current steering angle command and speed command, the steering angle command is input to the steering mechanism to control the steering mechanism to perform steering through the steering angle command; the speed command is input to the driving mechanism to control the driving mechanism to perform driving or braking through the speed command.

[0070] For example, Figure 4 Schematic diagram of the process of the automatic parking control method provided in this application Figure 2 ,like Figure 4 As shown, in the process of automatic parking based on the reference path, the heading deviation, lateral deviation, longitudinal deviation and speed deviation are determined according to the real-time data of the vehicle and the reference path, and the heading deviation is processed by a linear control strategy to obtain a heading steering angle instruction; the heading deviation, lateral deviation and longitudinal deviation are processed by a nonlinear control strategy to obtain a lateral steering angle instruction, and a steering angle instruction is determined according to the heading steering angle instruction and the lateral steering angle instruction, and the speed deviation is processed according to the linear control strategy to obtain a speed instruction; the steering angle instruction is input to the steering mechanism to control the steering mechanism to perform steering through the steering angle instruction; and the speed instruction is input to the driving mechanism to control the driving mechanism to perform driving or braking through the speed instruction.

[0071] S105 : When it is determined based on the target distance between the vehicle and the parking area that the parking completion condition is met, stop the automatic parking.

[0072] Optionally, during the automatic parking process, the vehicle can determine the vehicle's position using sensor data and, based on that position, detect a target distance between the vehicle and the parking area. When the target distance falls within a first preset distance interval, the parking completion condition is determined to be met, and the automatic parking function is terminated, exiting. The target distance falling within the first preset distance interval indicates that the distance between the vehicle and the parking area is very close, and the vehicle can be considered to have entered the parking area. The first preset distance interval can be set based on actual needs.

[0073] Optionally, during the automatic parking process, the vehicle side continuously detects the target distance between the vehicle and the parking area. When the target distance falls within a first preset distance interval, the distance between the vehicle and the obstacle is detected through sensor data. If the distance between the vehicle and the obstacle is greater than a preset safety distance threshold, it is determined that the vehicle is parked in the parking area, and then the automatic parking is stopped and the automatic parking function is exited.

[0074] Optionally, when the target distance does not belong to the first preset distance interval, it means that the distance between the vehicle and the parking area is large, and the process from S102 to S104 needs to be continued.

[0075] Optionally, after stopping automatic parking, the vehicle side records the relevant data of automatic parking and uploads it to the cloud server. The relevant data of automatic parking includes but is not limited to: parking time, location, environmental information and other data; uploading the relevant data of automatic parking to the cloud server allows users to view the parking records through the vehicle side or mobile terminal.

[0076] After stopping automatic parking, the user can also be notified of the completion of parking in a variety of ways; optionally, the parking completion information, the relative position between the vehicle's parking position and surrounding obstacles, and the distance between the vehicle and surrounding obstacles can be displayed on the vehicle-side display, so that the user can understand the parking status of the vehicle.

[0077] Optionally, a voice prompt is issued through the vehicle audio system. For example, the voice prompt may be: "Parking is successful, you are now parked in the designated area", "You are 30 cm away from surrounding obstacles, safe".

[0078] Optionally, a parking completion notification is pushed to a vehicle management application on the mobile terminal so that the user can view the parking status on the vehicle management application.

[0079] In actual application, after parking is completed, the user can fine-tune the vehicle position; specifically, the vehicle side responds to the trigger operation for the reverse fine-tuning control or the forward fine-tuning control, and moves the vehicle using preset fine-tuning parameters, so that the vehicle moves back or forward a preset fine-tuning distance; the reverse fine-tuning control and the forward fine-tuning control can be controls displayed on the vehicle side display screen, or they can be controls displayed on the mobile terminal.

[0080] After completing automatic parking, the user can also evaluate the automatic parking process; for example, the vehicle sends an evaluation prompt, the user clicks the evaluation prompt, enters the evaluation page, and evaluates the automatic process through the evaluation page (for example, selects an evaluation tag, determines a satisfaction score, enters suggestions or opinions), and submits the evaluation content to the cloud server.

[0081] The automatic parking control method proposed in the embodiment of the present application responds to the automatic parking instruction, identifies the parking area based on the sensor data, and performs path planning based on the parking area to obtain a reference path. In the process of automatic parking based on the reference path, a deviation set is determined according to the real-time data of the vehicle and the reference path, and the deviation set includes a heading deviation, a lateral deviation, a longitudinal deviation and a speed deviation; a linear control strategy is used to process the heading deviation to obtain a heading steering angle instruction; a nonlinear control strategy is used to fuse the heading deviation, the lateral deviation and the longitudinal deviation to obtain a lateral steering angle instruction; and a heading steering angle instruction is obtained according to the heading steering angle instruction and the lateral steering angle instruction. The steering angle command determines the steering angle command; the speed deviation is processed according to the linear control strategy to obtain the speed command, and the steering angle and speed of the vehicle are adjusted based on the steering angle command and the speed command; in the parking scenario, the nonlinear control strategy enables the automatic parking to have a strong path following capability, and the linear control strategy can respond quickly and eliminate steady-state errors. In the process of parking according to the reference path, the embodiment of the present application combines the nonlinear control strategy with the linear control strategy, which can quickly and stably correct the deviation, accurately achieve path tracking, and provide smooth steering control, thereby improving the accuracy and robustness of automatic parking.

[0082] In some embodiments, a steering angle command is determined based on a heading steering angle command and a lateral steering angle command, including: determining a target speed deviation interval corresponding to the speed deviation in a preset speed deviation interval; obtaining a preset linear weight and a preset nonlinear weight corresponding to the target speed deviation interval; and performing weighted summation of the heading steering angle command and the lateral steering angle command based on the preset linear weight and the preset nonlinear weight to obtain a steering angle command.

[0083] Among them, since the heading steering angle command is obtained by processing the heading deviation using a linear control strategy, the preset linear weight is used to indicate the importance of the heading steering angle command; since the lateral steering angle command is obtained by processing the heading deviation, lateral deviation and longitudinal deviation using a nonlinear control strategy, the preset nonlinear weight is used to indicate the importance of the lateral steering angle command.

[0084] Specifically, a plurality of preset speed deviation intervals are obtained, and the speed deviation is matched with the plurality of preset speed deviation intervals to determine the preset speed deviation interval to which the speed deviation belongs, and the preset speed deviation interval to which the speed deviation belongs is used as the target speed deviation interval; and a preset linear weight and a preset nonlinear weight corresponding to the target speed deviation interval are obtained.

[0085] The preset linear weight is used as the weight of the heading steering angle instruction, and the preset nonlinear weight is used as the weight of the lateral steering angle. The heading steering angle instruction and the lateral steering angle instruction are weighted and summed by the preset linear weight and the preset nonlinear weight to obtain the steering angle instruction.

[0086] It should be noted that, taking the first preset speed deviation interval and the second preset speed deviation interval included in multiple preset speed deviation intervals as an example, when the speed deviation contained in the first preset speed deviation interval is greater than the speed deviation contained in the second preset speed deviation interval, the preset linear weight corresponding to the first preset speed deviation interval is greater than the preset linear weight corresponding to the second preset speed deviation interval, and the preset nonlinear weight corresponding to the first preset speed deviation interval is less than the preset nonlinear weight corresponding to the second preset speed deviation interval.

[0087] That is to say, when the speed deviation is large (for example, the speed deviation belonging to the first preset speed deviation interval is greater than the speed deviation belonging to the second preset speed deviation interval), the weight corresponding to the heading steering angle instruction is larger (the preset linear weight corresponding to the first preset speed deviation interval is greater than the preset linear weight corresponding to the second preset speed deviation interval), and the weight corresponding to the lateral steering angle instruction is smaller (the preset nonlinear weight corresponding to the first preset speed deviation interval is smaller than the preset nonlinear weight corresponding to the second preset speed deviation interval); conversely, when the speed deviation is small, the weight corresponding to the heading steering angle instruction is smaller, and the weight corresponding to the lateral steering angle instruction is larger.

[0088] In this way, when the speed deviation is large, the focus is on correcting the heading deviation to avoid excessive correction of the lateral deviation, which will cause the vehicle's lateral control to become unstable. When the speed deviation is small, the focus is on correcting the lateral deviation, which can accurately correct the vehicle position and achieve more accurate path tracking.

[0089] In the above embodiment, the preset linear weight and the preset nonlinear weight are determined through the speed deviation, so that the contribution degree of the heading steering angle command obtained based on the linear control strategy to the steering angle command, and the contribution degree of the lateral steering angle command obtained based on the nonlinear control strategy to the steering angle command, are related to the speed deviation, so that when the speed deviation of the vehicle is large, the stability of the lateral correction can be guaranteed, and when the speed deviation is small, more accurate path tracking can be achieved.

[0090] In some embodiments, a nonlinear control strategy is used to fuse the heading deviation, lateral deviation and longitudinal deviation to obtain a lateral steering angle instruction, including: determining an adaptive gain based on the longitudinal deviation and a preset gain; and using a nonlinear control strategy to fuse the adaptive gain, lateral deviation and heading deviation to obtain a lateral steering angle instruction.

[0091] The preset gain is used to control the correction strength of the lateral error. The larger the preset gain is, the more sensitive the response is.

[0092] Specifically, a preset gain, a gain adjustment factor, and a distance sensitivity coefficient are obtained, a gain adjustment coefficient is determined according to the longitudinal deviation, the gain adjustment factor, and the distance sensitivity coefficient, and an adaptive gain is determined according to the gain adjustment coefficient and the preset gain; illustratively, as shown in formula (1).

[0093] Formula (1): ;

[0094] in, is the adaptive gain, is the preset gain, is the gain adjustment factor, is the distance sensitivity coefficient, is the longitudinal deviation.

[0095] Specifically, a nonlinear control strategy is used to fuse the adaptive gain, lateral deviation, and heading deviation to obtain a lateral steering angle command, including normalizing the lateral deviation according to the vehicle speed and adjusting the weight through the adaptive gain to obtain a normalized lateral deviation; nonlinearly processing the normalized lateral deviation to obtain a candidate steering angle command, and determining the lateral steering angle command based on the heading deviation and the candidate steering angle command; illustratively, as shown in formula (2).

[0096] Formula (2): ;

[0097] in, is the lateral steering angle command, is the heading deviation, is the adaptive gain, is the lateral deviation, is the current vehicle speed, It is a preset parameter.

[0098] In the above embodiment, the preset gain is adjusted by the longitudinal deviation to obtain an adaptive gain, and the adaptive gain is used as a weight for normalizing the lateral deviation. The adaptive gain determined by the longitudinal deviation improves the balance between sensitivity and stability during the automatic parking control process, and more accurate path tracking can be achieved.

[0099] In some embodiments, the steering angle and speed of the vehicle are adjusted based on the steering angle command and the speed command, including: determining whether the vehicle is in a speed adjustment area based on a target distance between the vehicle and the parking area; when the vehicle is not in the speed adjustment area, adjusting the steering angle and speed of the vehicle based on the steering angle command and the speed command.

[0100] Specifically, the real-time distance between the vehicle and the parking area is used as the target distance. If the target distance does not fall within the second preset distance interval, it is determined that the vehicle is not in the speed adjustment area. If the target distance falls within the second preset distance interval, it is determined that the vehicle is in the speed adjustment area. The second preset distance interval can be set according to actual needs.

[0101] If the vehicle is in the speed adjustment area, it means that the vehicle is close to the parking area; if the vehicle is not in the speed adjustment area, it means that the vehicle is far away from the parking area; when the vehicle is not in the speed adjustment area, the steering angle and speed of the vehicle are adjusted based on the steering angle command and the speed command.

[0102] It should be noted that, while the previous description determined that the vehicle had parked in the parking area by determining that the target distance belonged to the first preset distance interval, this embodiment determines that the vehicle is approaching the parking area by determining that the target distance belongs to the second preset distance interval. The first preset distance interval is included in the second preset distance interval. When the target distance belongs to the first preset distance interval, it indicates that the vehicle has arrived near the end point of the reference path. When the target distance belongs to the second preset distance interval, it indicates that the vehicle may still be some distance away from the end point of the reference path. For example, the first preset distance interval may be [0, 5 cm], and the second preset distance interval may be [0, 1 m].

[0103] Optionally, in addition to determining whether the vehicle is in the speed regulation area based on the target distance between the vehicle and the parking area, whether the vehicle is in the speed regulation area can also be determined based on the longitudinal deviation; if the longitudinal deviation is less than the longitudinal deviation threshold, it is determined that the vehicle is in the speed regulation area.

[0104] In some embodiments, the automatic parking control method further includes: determining whether the vehicle is in a speed adjustment area based on a target distance between the vehicle and the parking area, or a minimum distance between the vehicle and an obstacle; and when the vehicle is not in the speed adjustment area, adjusting the steering angle and vehicle speed of the vehicle based on the steering angle command and the speed command.

[0105] Specifically, obstacles around the vehicle are determined through sensor data, and the initial distance between the vehicle and the obstacle is determined based on the current position of the vehicle and the position of the obstacle; when there are multiple obstacles around the vehicle, the initial distances between the vehicle and the multiple obstacles are determined respectively, and the minimum distance is determined among the multiple initial distances.

[0106] If the target distance or the minimum distance falls within the preset speed regulation range, the vehicle is determined to be in the speed regulation area; if the target distance falls within the preset speed regulation range, it means that the vehicle is approaching the parking area, and if the minimum distance falls within the preset speed regulation range, it means that the vehicle is approaching an obstacle.

[0107] If both the target distance and the minimum distance do not fall within the preset speed adjustment range, it is determined that the vehicle is not in the speed adjustment area, which means that the vehicle is far away from the parking area and far away from the obstacle.

[0108] In the above embodiment, since the speed command is obtained by linearly controlling the speed deviation, the speed command prioritizes efficiency, which can more quickly shorten the distance between the vehicle and the parking area. When the vehicle is far away from the parking area, there are usually fewer obstacles around the vehicle. Adjusting the steering angle and speed of the vehicle based on the steering angle command and the speed command can improve the efficiency of automatic parking.

[0109] In some embodiments, the automatic parking control method further includes: when the vehicle is in a speed adjustment area, adjusting the speed instruction according to the target distance and the lateral deviation included in the deviation set to obtain a target speed instruction; and adjusting the steering angle and vehicle speed of the vehicle based on the steering angle instruction and the target speed instruction.

[0110] Specifically, the vehicle is in the speed adjustment area, which means that the vehicle is approaching the parking area. In actual applications, when the vehicle approaches the parking area, there may be many obstacles around the vehicle, for example, there are flower beds, pillars and other obstacles near the parking area.

[0111] To avoid collision risks during automatic parking, when the vehicle approaches the parking area, the speed command is adjusted through lateral deviation to obtain the target speed command to reduce the parking speed. The steering angle and speed of the vehicle are adjusted through the steering angle command and the target speed command.

[0112] For example, Figure 5 As shown, the speed regulation area may be a circular area with the center of the parking area as the center, and the distance between a vehicle in the speed regulation area and the parking area is smaller than the distance between a vehicle not in the speed regulation area and the parking area.

[0113] After obtaining the steering angle command and speed command, determine whether the vehicle is in the speed adjustment area. For example, if the current position of the vehicle is p1 and it is determined that the vehicle is not in the speed adjustment area, adjust the steering angle and speed of the vehicle based on the steering angle command and speed command.

[0114] For example, the current position of the vehicle is p2, and it is determined that the vehicle is in the speed adjustment area. The speed command is adjusted by the lateral deviation to obtain the target speed command. Based on the steering angle command and the target speed command, the steering angle and speed of the vehicle are adjusted.

[0115] It should be noted that when the vehicle is in the speed regulation area, the steering angle and speed of the vehicle are adjusted based on the steering angle command and the target speed command; when the vehicle is not in the speed regulation area, the steering angle and speed of the vehicle are adjusted based on the steering angle command and the speed command; this means that whether the "speed command" obtained through the embodiment of S103 is adjusted is determined by whether the vehicle is in the speed regulation area; the "steering angle commands" in the above two places are both obtained through the embodiment of S103, and whether the vehicle is in the speed regulation area has no effect on the steering angle command.

[0116] Optionally, the speed instruction is adjusted by the lateral deviation to obtain a target speed instruction, including: determining the absolute value of the lateral deviation, determining a fine-tuning coefficient based on the absolute value of the lateral deviation and a lateral deviation threshold, calculating the product between the fine-tuning coefficient and the speed instruction, and using the difference between the speed instruction and the product as the target speed instruction.

[0117] Optionally, the speed instruction is adjusted by the lateral deviation to obtain a target speed instruction, including: performing proportional control and differential control on the lateral deviation to obtain a proportional lateral deviation and a differential lateral deviation, using the sum of the proportional lateral deviation and the differential lateral deviation as the speed adjustment value, using the difference between the speed instruction and the speed adjustment value as a candidate speed instruction, determining a candidate speed deviation based on the candidate speed instruction and a reference speed, and then processing the candidate speed deviation according to a speed PID controller to obtain a target speed instruction.

[0118] Optionally, the speed instruction is adjusted according to the target distance and the lateral deviation included in the deviation set to obtain the target speed instruction, including: determining the compensation deviation according to the lateral deviation included in the deviation set; determining the distance compensation coefficient according to the target distance and a preset distance threshold; and determining the target speed instruction according to the compensation deviation, the distance compensation coefficient and the speed instruction.

[0119] Specifically, the lateral deviation is input into the compensation function to obtain the compensation deviation; the difference between the target distance and the preset distance threshold is calculated, and the ratio between the difference and the speed regulation area threshold is determined to obtain a candidate compensation coefficient. The minimum value between the candidate compensation coefficient and 1 is used as the distance compensation coefficient, and the product of the compensation deviation, the distance compensation coefficient and the speed command is calculated to obtain the target speed command; wherein the speed regulation area threshold represents the distance range of the speed regulation area. For example, when the speed regulation area is a circular area with the center of the parking area as the center, the speed regulation area threshold can be the radius of the speed regulation area.

[0120] For example, as shown in formula (3).

[0121] Formula (3): ;

[0122] in, is the target speed command, is the speed command, is the target distance, Preset distance threshold, is the speed regulation area threshold, is the lateral deviation, is the compensation deviation.

[0123] Among them, compensation deviation It can be determined by formula (4).

[0124] Formula (4): ;in, is the compensation deviation, is the preset compensation coefficient, is the lateral deviation.

[0125] In the above embodiment, when the vehicle is not in the speed regulation area, the parking speed can be higher. When the vehicle is in the speed regulation area, the speed command is adjusted through lateral deviation to reduce the parking speed, so that the vehicle moves at a low speed in the final stage of automatic parking to avoid the risk of collision.

[0126] In some embodiments, adjusting the steering angle and speed of the vehicle based on the steering angle command and the speed command also includes: determining whether the lateral deviation is greater than a preset lateral threshold, and determining whether the heading deviation is greater than a preset heading threshold; if the lateral deviation is greater than the preset lateral threshold, or the heading deviation is greater than the preset heading threshold, adjusting the steering angle command to obtain a target steering angle command; adjusting the steering angle and speed of the vehicle according to the target steering angle command and the speed command.

[0127] Specifically, when the lateral deviation is greater than a preset lateral threshold, the steering angle command can be adjusted by determining a first difference between the lateral deviation and the preset lateral threshold, determining a first target difference interval to which the first difference belongs among multiple preset difference intervals, using the preset adjustment coefficient corresponding to the first target difference interval as the first adjustment coefficient corresponding to the first difference, calculating the product between the first adjustment coefficient and the steering angle command, and obtaining the target steering angle command.

[0128] When the heading deviation is greater than a preset heading threshold, the steering angle command can be adjusted by determining a second difference between the heading deviation and the preset heading threshold, determining a second target difference interval to which the second difference belongs among multiple preset difference intervals, using the preset adjustment coefficient corresponding to the second target difference interval as the second adjustment coefficient corresponding to the second difference, calculating the product between the second adjustment coefficient and the steering angle command, and obtaining the target steering angle command.

[0129] In the above embodiment, the steering angle command is adjusted according to the lateral deviation and the heading deviation, so that an excessive lateral deviation or heading deviation can be corrected by the steering angle command, thereby improving the accuracy of path tracking.

[0130] In some embodiments, the automatic parking control method further includes: during the process of automatic parking based on the reference path, performing target detection through real-time sensor data to obtain position information of dynamic obstacles; pausing the automatic parking process when it is determined that the dynamic obstacle belongs to the parking area based on the position information of the dynamic obstacle; re-identifying the target parking area based on the real-time sensor data; updating the parking area to the target parking area, returning to the step of performing path planning based on the parking area to obtain the reference path, and continuing execution.

[0131] Specifically, during automatic parking based on a reference path, the vehicle obtains real-time sensor data and detects dynamic obstacles through the real-time sensor data. By comparing the dynamic obstacles in consecutive frames and using a Kalman filter or target tracking algorithm, the motion trajectory of the dynamic obstacle can be determined. Based on the motion trajectory of the dynamic obstacle, the position information of the dynamic obstacle can be determined.

[0132] When it is determined that the dynamic obstacle is within the parking area based on the position information of the dynamic obstacle, the automatic parking process is suspended, and the process of S101 is executed based on the real-time sensor data to re-identify the target parking area. The parking area is replaced with the re-identified target parking area, and the path planning is re-performed to obtain a reference path. The process is continued from S102 to park the vehicle in the re-determined target parking area.

[0133] Optionally, when it is determined based on the position information of the dynamic obstacle that the distance between the vehicle and the dynamic obstacle is less than a preset safety distance, a braking instruction is executed.

[0134] Optionally, during the process of automatic parking based on the reference path, when both the lateral deviation and the heading deviation exceed their respective deviation thresholds, the automatic parking process is paused; the target parking area is re-identified based on the real-time sensor data; the parking area is updated to the target parking area, and the process returns to the step of performing path planning based on the parking area to obtain the reference path and continue execution.

[0135] The heading deviation exceeds the corresponding deviation threshold, which can be less than or equal to -3°, or greater than or equal to The lateral deviation exceeding the corresponding deviation threshold may be that the lateral deviation is less than or equal to -5 cm, or greater than or equal to 5 cm.

[0136] Optionally, during the process of automatic parking based on the reference path, if there is a fault in the vehicle unit, a fault handling program is triggered, the automatic parking is suspended through the fault handling program, and the driver is notified to take over.

[0137] In the above embodiment, when a dynamic obstacle is detected in the parking area, the parking area is redefined to avoid collision accidents caused by the dynamic obstacle, thereby ensuring parking safety.

[0138] In some embodiments, the sensor data includes surround-view images, distance data, and laser point cloud data; identifying a parking area based on the sensor data includes: determining a first candidate parking area based on the surround-view images and the distance data; the first candidate parking area does not include any obstacles; selecting a second candidate parking area in the first candidate parking area whose size meets the parking conditions; determining the slope and flatness of the second candidate parking area based on the laser point cloud data; and selecting a parking area in the second candidate parking area based on the slope and flatness.

[0139] It should be noted that this embodiment is an embodiment for determining a parking area in a ground scene without parking space lines; Figure 6 As shown, ground scenes without parking space lines include: open spaces without obvious parking space lines, slopes and gravel surfaces, and urban streets without obvious parking space lines.

[0140] Specifically, images are acquired through multiple on-board cameras with different perspectives, and the multiple images are stitched together to obtain a surround view image. Distance data is acquired through ultrasonic radar, and point cloud data is acquired through lidar. Time synchronization is achieved among the surround view image, distance data, and point cloud data through the National Time Protocol (NTP) or other clock synchronization mechanisms. The timestamps of the surround view image, distance data, and point cloud data used to ensure the parking area are consistent.

[0141] For example, Figure 7 As shown in FIG, the surround view image is subjected to image denoising and enhancement processing to obtain a processed surround view image; for example, denoising algorithms such as image Gaussian filtering and wavelet transform are used to denoise the surround view image to eliminate noise in the surround view image; for example, the contrast and brightness of the surround view image are enhanced through methods such as histogram equalization and gamma correction, so that image features under different lighting conditions are more obvious.

[0142] The distance data is corrected to obtain the corrected distance data. The correction can eliminate measurement errors and compensate for influencing factors such as ambient temperature and humidity. The point cloud data is filtered to obtain processed point cloud data. The filtering process can remove noise points, reduce the amount of calculation, and maintain key environmental characteristics.

[0143] A convolutional neural network is used to extract features from the processed surround view image to obtain a feature map. The convolutional neural network can automatically learn and extract image features through a multi-layer network structure and has strong generalization capabilities for complex visual data. The feature map is segmented to obtain open space areas and obstacle areas.

[0144] The size of the open space is determined based on the distance data. If the size is larger than a preset size, the open space is selected as a first candidate parking area. If the size of the first candidate parking area is larger than the preset size, it means that the length, width, and height of the first candidate parking area are sufficient to accommodate the vehicle.

[0145] A candidate distance between the first candidate parking area and the obstacle area is determined based on the distance data, and the first candidate parking area having a candidate distance greater than a reference distance is used as a second candidate parking area.

[0146] There can be multiple second candidate parking areas. For each second candidate parking area, a ground point set is extracted from the point cloud data, and the slope and flatness of the second candidate parking area are determined based on the ground point set. For example, the ground point set is processed using the least squares method to fit a plane, and the angle between the plane and the horizontal plane is calculated to obtain the slope. The ground point set is processed using a flatness calculation method to obtain the flatness. For example, the ground point set can be processed using the root mean square (RMS) flatness to obtain the flatness.

[0147] For each second candidate parking area, a parking grade of the second candidate parking area is determined from preset grades based on the slope interval and the flatness interval of the second candidate parking area. The preset grades include: safety grade, general risk grade, medium risk grade, and high risk grade. Among them, the safety grade is higher than the general risk grade; the general risk grade is higher than the medium risk grade; and the medium risk grade is higher than the high risk grade.

[0148] The second candidate parking area with the highest parking level among the multiple second candidate parking areas and not having a high-risk parking level is selected as the parking area. In other words, if the highest parking level among the multiple second candidate parking areas is a high-risk level, no suitable parking area can be selected from the multiple second candidate parking areas.

[0149] For example, the parking grade of the second candidate parking area is determined in preset grades according to the interval to which the slope of the second candidate parking area belongs and the interval to which the flatness belongs. For details, see Table 1.

[0150] Table 1

[0151]

[0152] When the parking area is rated as medium risk, activate suspension level control or chassis adaptive suspension, and avoid high-slope locations during route planning. When the parking area is rated as medium risk, activate suspension level control or chassis adaptive suspension, and avoid high-slope locations during route planning, and prompt the user that the slope is large and please park with caution.

[0153] Optionally, identifying parking areas based on sensor data can be achieved through a recognition model, where the sensor data is input into the recognition model, and the open space area and obstacle area are determined through the recognition model; the recognition model includes a feature extraction layer and a classification layer, where the feature extraction layer is used to extract feature maps, and the classification layer is used to identify the open space area and obstacle area.

[0154] The recognition model is trained using sample surround view images, open space labels, and obstacle labels to train an initial neural network model. The open space labels are obtained by annotating the areas of open space available for parking in the sample surround view images, while the obstacle labels are obtained by annotating the areas where obstacles are located in the sample surround view images.

[0155] The sample surround view image is input into the initial neural network model to obtain the predicted open space area and preset obstacle area. The loss value is calculated through the cross entropy loss function based on the predicted open space area, open space area label, predicted obstacle area and obstacle area label. The parameters of the initial neural network model are adjusted according to the loss value. It is trained iteratively until the initial neural network model converges. The converged initial neural network model is used as the recognition model.

[0156] In the above embodiment, by using surround view images, distance data and point cloud data, parking areas can be determined not only in parking scenarios with parking space lines, but also in open areas without parking space lines. This makes automatic parking applicable to various parking scenarios and improves the generalization capability of automatic parking application scenarios.

[0157] The automatic parking control method proposed in the embodiment of the present application responds to the automatic parking instruction, identifies the parking area based on the sensor data, and performs path planning based on the parking area to obtain a reference path. In the process of automatic parking based on the reference path, a deviation set is determined according to the real-time data of the vehicle and the reference path, and the deviation set includes a heading deviation, a lateral deviation, a longitudinal deviation and a speed deviation; a linear control strategy is used to process the heading deviation to obtain a heading steering angle instruction; a nonlinear control strategy is used to fuse the heading deviation, the lateral deviation and the longitudinal deviation to obtain a lateral steering angle instruction; and a heading steering angle instruction is obtained according to the heading steering angle instruction and the lateral steering angle instruction. The steering angle command determines the steering angle command; the speed deviation is processed according to the linear control strategy to obtain the speed command, and the steering angle and speed of the vehicle are adjusted based on the steering angle command and the speed command; in the parking scenario, the nonlinear control strategy enables the automatic parking to have a strong path following capability, and the linear control strategy can respond quickly and eliminate steady-state errors. In the process of parking according to the reference path, the embodiment of the present application combines the nonlinear control strategy with the linear control strategy, which can quickly and stably correct the deviation, accurately achieve path tracking, and provide smooth steering control, thereby improving the accuracy and robustness of automatic parking.

[0158] It should be understood that, although the steps in the flowcharts of the above embodiments are shown in sequence as indicated by the arrows, these steps are not necessarily performed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps can be performed in other orders. Moreover, at least a portion of the steps in the flowcharts of the above embodiments may include multiple steps or multiple stages, and these steps or stages are not necessarily performed at the same time, but can be performed at different times. The execution order of these steps or stages is not necessarily to be performed in sequence, but can be performed in turn or alternately with other steps or at least a portion of steps or stages in other steps.

[0159] Figure 8 The schematic diagram of the automatic parking control device provided in this application is as follows: Figure 8 As shown, the automatic parking control device 80 provided in this embodiment includes:

[0160] a path planning module 801 for identifying a parking area based on sensor data in response to an automatic parking instruction, and performing path planning based on the parking area to obtain a reference path;

[0161] Deviation determination module 802, configured to determine a deviation set based on the vehicle's real-time data and the reference path during automatic parking based on the reference path; the deviation set includes heading deviation, lateral deviation, longitudinal deviation, and speed deviation;

[0162] The command determination module 803 is configured to process the heading deviation using a linear control strategy to obtain a heading steering angle command, fuse the heading deviation, lateral deviation, and longitudinal deviation using a nonlinear control strategy to obtain a lateral steering angle command, determine a steering angle command based on the heading steering angle command and the lateral steering angle command, and process the speed deviation using a linear control strategy to obtain a speed command.

[0163] A dynamic adjustment module 804 is used to adjust the steering angle and speed of the vehicle based on the steering angle command and the speed command;

[0164] The parking stop module 805 is configured to stop the automatic parking when it is determined that a parking completion condition is met based on a target distance between the vehicle and the parking area.

[0165] In one possible implementation, the instruction determination module 803 is used to determine a target speed deviation interval corresponding to the speed deviation in a preset speed deviation interval; obtain a preset linear weight and a preset nonlinear weight corresponding to the target speed deviation interval; and perform weighted summation on the heading steering angle instruction and the lateral steering angle instruction according to the preset linear weight and the preset nonlinear weight to obtain a steering angle instruction.

[0166] In one possible implementation, the instruction determination module 803 is configured to determine an adaptive gain based on the longitudinal deviation and a preset gain; and adopt a nonlinear control strategy to fuse the adaptive gain, the lateral deviation, and the heading deviation to obtain a lateral steering angle instruction.

[0167] In one possible implementation, the dynamic adjustment module 804 is configured to determine whether the vehicle is in a speed adjustment area based on a target distance between the vehicle and the parking area; and when the vehicle is not in the speed adjustment area, adjust the steering angle and vehicle speed of the vehicle based on the steering angle command and the speed command.

[0168] In one possible implementation, the dynamic adjustment module 804 is further used to adjust the speed instruction according to the target distance and the lateral deviation included in the deviation set to obtain a target speed instruction when the vehicle is not in the speed adjustment area; and adjust the steering angle and vehicle speed of the vehicle based on the steering angle instruction and the target speed instruction.

[0169] In one possible implementation, the dynamic adjustment module 804 is further configured to determine a compensation deviation based on the lateral deviation included in the deviation set; determine a distance compensation coefficient based on the target distance and a preset distance threshold; and determine a target speed instruction based on the compensation deviation, the distance compensation coefficient, and the speed instruction.

[0170] In one possible implementation, the dynamic adjustment module 804 is further configured to, during automatic parking based on a reference path, perform target detection using real-time sensor data to obtain location information of dynamic obstacles; suspend the automatic parking process when it is determined based on the location information of the dynamic obstacle that the dynamic obstacle belongs to the parking area; re-identify the target parking area based on the real-time sensor data; update the parking area to the target parking area, return to the step of performing path planning based on the parking area to obtain the reference path, and continue execution.

[0171] In one possible implementation, the sensor data includes surround view images, distance data, and laser point cloud data; the path planning module 801 is configured to determine a first candidate parking area based on the surround view images and the distance data; the first candidate parking area does not include any obstacles; a second candidate parking area is selected from the first candidate parking area, the size of which meets the parking conditions; the slope and flatness of the second candidate parking area are determined based on the laser point cloud data; and a parking area is selected from the second candidate parking area based on the slope and flatness.

[0172] The automatic parking control device provided in this embodiment can execute the automatic parking control method provided in the above method embodiment. Its implementation principle and technical effects are similar, and are not described in detail in this embodiment.

[0173] Figure 9 This is a schematic diagram of the structure of the electronic device provided in this application. Figure 9 As shown, the electronic device 90 provided in this embodiment includes: at least one processor 901 and a memory 902. Optionally, the device 90 further includes a communication component 903. The processor 901, the memory 902 and the communication component 903 are connected via a bus.

[0174] During the specific implementation process, at least one processor 901 executes the computer-executable instructions stored in the memory 902, so that the at least one processor 901 performs the above method.

[0175] The specific implementation process of the processor 901 can be found in the above method embodiment. Its implementation principle and technical effects are similar and will not be repeated here in this embodiment.

[0176] In the above embodiments, it should be understood that the processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASICs), etc. A general-purpose processor may be a microprocessor or any conventional processor. The steps of the method disclosed in the present invention may be directly executed by a hardware processor or by a combination of hardware and software modules within the processor.

[0177] The memory may include random access memory (RAM) and may also include non-volatile memory (NVM), such as at least one disk storage.

[0178] A bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus. Buses can be categorized as address buses, data buses, and control buses. For ease of illustration, the buses in the drawings of this application are not limited to just one bus or just one type of bus.

[0179] The present application also provides a computer program product, including a computer program, which implements the above method when executed by a processor.

[0180] The present application also provides a computer-readable storage medium, in which computer-executable instructions are stored. When a processor executes the computer-executable instructions, the above method is implemented.

[0181] The readable storage medium may be implemented by any type of volatile or non-volatile memory device, or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The readable storage medium may be any available medium that can be accessed by a general-purpose or special-purpose computer.

[0182] An exemplary readable storage medium is coupled to a processor so that the processor can read information from the readable storage medium and write information to the readable storage medium. Of course, the readable storage medium can also be an integral part of the processor. The processor and the readable storage medium can be located in an application specific integrated circuit (ASIC). Of course, the processor and the readable storage medium can also exist in the device as discrete components.

[0183] The division of units is merely a logical functional division; actual implementations may employ alternative divisions, such as combining or integrating multiple units or components into another system, or omitting or disabling certain features. Furthermore, any direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection between devices or units, either through an interface, electrical, mechanical, or other means.

[0184] Units described as separate components may or may not be physically separate, and components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0185] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0186] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the various embodiments of the method of the present invention. The aforementioned storage medium includes various media that can store program code, such as USB flash drives, mobile hard drives, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical disks.

[0187] Those skilled in the art will appreciate that all or part of the steps in the above-described method embodiments can be implemented using hardware associated with program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0188] Finally, it should be noted that those skilled in the art will readily identify other embodiments of the present invention after considering the specification and practicing the invention disclosed herein. The present invention is intended to cover any variations, uses, or adaptations of the present invention that follow the general principles of the present invention and include common knowledge or customary techniques in the art not disclosed herein. The present invention is not limited to the precise structure described above and illustrated in the accompanying drawings, and various modifications and variations may be made without departing from the scope thereof. The scope of the present invention is limited solely by the appended claims.

Claims

1. An automatic parking control method, characterized in that: include: In response to an automatic parking instruction, identifying a parking area based on sensor data, and performing path planning based on the parking area to obtain a reference path; During automatic parking based on the reference path, determining a deviation set based on the vehicle real-time data and the reference path; the deviation set includes a heading deviation, a lateral deviation, a longitudinal deviation, and a speed deviation; The heading deviation is processed using a linear control strategy to obtain a heading steering angle instruction, the heading deviation, the lateral deviation, and the longitudinal deviation are integrated using a nonlinear control strategy to obtain a lateral steering angle instruction, a steering angle instruction is determined based on the heading steering angle instruction and the lateral steering angle instruction, and the speed deviation is processed using a linear control strategy to obtain a speed instruction; adjusting the steering angle and the vehicle speed of the vehicle based on the steering angle command and the speed command; When it is determined that a parking completion condition is satisfied based on a target distance between the vehicle and the parking area, automatic parking is stopped.

2. The method according to claim 1, characterized in that The determining of the steering angle command according to the heading steering angle command and the lateral steering angle command comprises: Determining a target speed deviation interval corresponding to the speed deviation in a preset speed deviation interval; Obtaining a preset linear weight and a preset nonlinear weight corresponding to the target speed deviation interval; The heading steering angle instruction and the lateral steering angle instruction are weightedly summed according to the preset linear weight and the preset nonlinear weight to obtain a steering angle instruction.

3. The method according to claim 1, characterized in that The adopting of a nonlinear control strategy to fuse the heading deviation, the lateral deviation, and the longitudinal deviation to obtain a lateral steering angle instruction includes: determining an adaptive gain according to the longitudinal deviation and a preset gain; A nonlinear control strategy is adopted to fuse the adaptive gain, the lateral deviation and the heading deviation to obtain a lateral steering angle instruction.

4. The method according to claim 1, wherein The adjusting the steering angle and the vehicle speed of the vehicle based on the steering angle command and the speed command includes: determining whether the vehicle is in a speed regulation area based on a target distance between the vehicle and the parking area; When the vehicle is not in the speed adjustment area, the steering angle and the vehicle speed of the vehicle are adjusted based on the steering angle command and the speed command.

5. The method according to claim 4, characterized in that The method further comprises: When the vehicle is in the speed adjustment area, adjusting the speed instruction according to the target distance and the lateral deviation included in the deviation set to obtain a target speed instruction; Based on the steering angle command and the target speed command, the steering angle and the vehicle speed of the vehicle are adjusted.

6. The method according to claim 5, characterized in that The adjusting the speed instruction according to the target distance and the lateral deviation included in the deviation set to obtain a target speed instruction includes: determining a compensation deviation according to the lateral deviation included in the deviation set; determining a distance compensation coefficient according to the target distance and a preset distance threshold; A target speed command is determined according to the compensation deviation, the distance compensation coefficient, and the speed command.

7. The method according to any one of claims 1 to 6, characterized in that The method further comprises: During the automatic parking process based on the reference path, target detection is performed using real-time sensor data to obtain position information of dynamic obstacles; pausing the automatic parking process when it is determined based on the position information of the dynamic obstacle that the dynamic obstacle belongs to the parking area; re-identifying a target parking area based on the real-time sensor data; The parking area is updated to the target parking area, and the process returns to the step of performing path planning based on the parking area to obtain a reference path, and continues to execute.

8. The method according to any one of claims 1 to 6, characterized in that The sensor data includes surround view images, distance data and laser point cloud data; The identifying of the parking area based on the sensor data includes: determining a first candidate parking area based on the surround view image and the distance data; the first candidate parking area does not include any obstacles; Selecting a second candidate parking area from the first candidate parking area, the size of which meets the parking condition; determining a slope and flatness of the second candidate parking area based on the laser point cloud data; A parking area is selected from the second candidate parking areas according to the slope and the flatness.

9. An automatic parking control device, characterized in that: The device comprises: a path planning module, configured to identify a parking area based on sensor data in response to an automatic parking instruction, and perform path planning based on the parking area to obtain a reference path; a deviation determination module, configured to determine a deviation set based on the vehicle's real-time data and the reference path during automatic parking based on the reference path; the deviation set comprising a heading deviation, a lateral deviation, a longitudinal deviation, and a speed deviation; an instruction determination module, configured to process the heading deviation using a linear control strategy to obtain a heading steering angle instruction, fuse the heading deviation, the lateral deviation, and the longitudinal deviation using a nonlinear control strategy to obtain a lateral steering angle instruction, determine a steering angle instruction based on the heading steering angle instruction and the lateral steering angle instruction, and process the speed deviation using a linear control strategy to obtain a speed instruction; a dynamic adjustment module, configured to adjust the steering angle and vehicle speed of the vehicle based on the steering angle command and the speed command; The parking stop module is configured to stop the automatic parking when it is determined that a parking completion condition is satisfied based on a target distance between the vehicle and the parking area.

10. An electronic device, characterized in that: include: a processor, and a memory communicatively connected to the processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the method according to any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer-executable instructions, which are used to implement the method according to any one of claims 1 to 8 when executed by a processor.

12. A computer program product, characterized in that The method comprises computer-executable instructions, which implement the method according to any one of claims 1 to 8 when the computer-executable instructions are executed by a processor.

Images