Parking path generation method and device

Abstract

This application provides a parking path generation method and device, which includes: determining a parking starting point corresponding to the vehicle's initial position information in a pre-generated dual-resolution raster map, and adding the parking starting point to a preset list of candidate points; popping out the lowest-cost node from the list of candidate points, and adding the lowest-cost node to a preset list of selected points; performing an RS curve search based on the nodes in the selected point list to obtain an RS curve search result; determining whether the RS curve search result passes obstacle collision detection; and generating a parking path based on the RS curve search result if the RS curve search result passes obstacle collision detection. This method and device can improve available parking scenarios, enhance the real-time performance of parking path planning, and improve the user experience when parking.

Description

Technical Field

[0001] The present application relates to the field of automobile technology, and in particular to a parking path generation method and device. Background Art

[0002] Currently, various manufacturers use the Hybrid A-Star algorithm to plan parking paths. However, this algorithm can easily fail in narrow parking spaces. Furthermore, its implementation requires certain map resolution, forcing it to make trade-offs between performance and speed. Furthermore, the algorithm inevitably suffers from sudden changes in curvature, which severely impacts the user experience during parking. Summary of the Invention

[0003] The purpose of the embodiments of the present application is to provide a parking path generation method and device that can improve the available parking scenarios, improve the real-time performance of parking path planning, and enhance the user experience when parking.

[0004] A first aspect of an embodiment of the present application provides a parking path generation method, comprising:

[0005] Determining a parking starting point corresponding to the initial position information of the vehicle in a pre-generated dual-resolution raster map, and adding the parking starting point to a preset list of candidate points;

[0006] Popping out the node with the lowest cost from the list of candidate points, and adding the node with the lowest cost to a preset list of selected points;

[0007] Performing an RS curve search based on the nodes in the selected point list to obtain an RS curve search result;

[0008] Determine whether the RS curve search result passes the obstacle collision detection;

[0009] When the RS curve search result passes the obstacle collision detection, a parking path is generated based on the RS curve search result.

[0010] Furthermore, the method further comprises:

[0011] When the RS curve search result fails the obstacle collision detection, performing node expansion based on a hybrid A-star algorithm in the dual-resolution grid map to obtain an expanded node;

[0012] The extended node is added to the list of candidate points, and the steps of popping out the node with the lowest cost from the list of candidate points and adding the node with the lowest cost to the preset list of selected points are triggered.

[0013] Furthermore, after the step of adding the extended node to the list of candidate points, the method further includes:

[0014] Detecting whether there is a candidate node in the candidate point list that is repeated with the extended node;

[0015] When the candidate node exists, calculating the cost of the extended node;

[0016] When the cost of the extended node is less than the cost of the node to be selected, the cost of the node to be selected is updated based on the cost of the extended node, and the node to be selected and the extended node are merged.

[0017] Furthermore, before the step of determining the parking starting point corresponding to the initial position information of the vehicle in the pre-generated dual-resolution grid map, the method further includes:

[0018] Obtaining parking space information, vehicle initial position information, and / or obstacle position information;

[0019] A dual-resolution grid map is generated based on the parking space information, the vehicle initial position information and / or the obstacle position information.

[0020] Furthermore, the method further comprises:

[0021] Determine the parking destination corresponding to the parking space to be parked in the pre-generated dual-resolution raster map:

[0022] Determining whether there are obstacles at the parking starting point and the parking end point;

[0023] When the obstacles do not exist at the parking starting point and the parking end point, the step of adding the parking starting point to a preset list of candidate points is triggered.

[0024] Furthermore, before the step of determining whether the RS curve search result passes the obstacle collision detection, the method further includes:

[0025] When a sudden change in curvature occurs in the RS curve search result, the RS curve search result is smoothed in combination with the properties of the clothoid curve; wherein, when the RS curve search result is a straight-curved curve type, the length of the straight line in the RS curve search result is greater than or equal to the length of the curve.

[0026] A second aspect of an embodiment of the present application provides a parking path generation device, the parking path generation device comprising:

[0027] a determining unit, configured to determine a parking starting point corresponding to the initial position information of the vehicle in a pre-generated dual-resolution raster map, and add the parking starting point to a preset list of candidate points;

[0028] An adding unit, configured to pop out a node with the lowest cost from the list of candidate points, and add the node with the lowest cost to a preset list of selected points;

[0029] A search unit, configured to perform an RS curve search based on the nodes in the selected point list to obtain an RS curve search result;

[0030] A judging unit, configured to judge whether the RS curve search result passes the obstacle collision detection;

[0031] A generating unit is configured to generate a parking path based on the RS curve search result when the RS curve search result passes the obstacle collision detection.

[0032] Furthermore, the parking path generating device further includes:

[0033] An expansion unit, configured to, when the RS curve search result fails the obstacle collision detection, perform node expansion in the dual-resolution grid map based on a hybrid A-star algorithm to obtain an expanded node;

[0034] The adding unit is further configured to add the extended node to the list of candidate points, and trigger the operation of popping out the node with the lowest cost from the list of candidate points, and adding the node with the lowest cost to the preset list of selected points.

[0035] Furthermore, the parking path generating device further includes:

[0036] a detection unit, configured to detect whether there is a candidate node in the candidate point list that is repeated with the extension node after the adding unit adds the extension node to the candidate point list;

[0037] a calculation unit, configured to calculate the cost of the extended node when the candidate node exists;

[0038] An updating unit is configured to update the cost of the node to be selected based on the cost of the extended node when the cost of the extended node is less than the cost of the node to be selected, and merge the node to be selected and the extended node.

[0039] Furthermore, the parking path generating device further includes:

[0040] an acquiring unit, configured to acquire information about a parking space to be parked, information about the vehicle's initial position, and / or information about an obstacle's position before the determining unit determines, in the pre-generated dual-resolution raster map, a parking starting point corresponding to the vehicle's initial position information;

[0041] The generating unit is further configured to generate a dual-resolution grid map based on the parking space information, the vehicle initial position information, and / or the obstacle position information.

[0042] Furthermore, the parking path generating device further includes:

[0043] The determining unit is further configured to determine a parking destination corresponding to the parking space information to be parked in the pre-generated dual-resolution grid map:

[0044] a judgment unit, configured to judge whether there are obstacles at the parking starting point and the parking end point;

[0045] The determining unit is further configured to add the parking starting point to a preset list of candidate points when no obstacles exist at the parking starting point and the parking end point.

[0046] Furthermore, the parking path generating device further includes:

[0047] a processing unit configured to smooth the RS curve search result by combining the properties of a clothoid curve before the judgment unit determines whether the RS curve search result passes the obstacle collision detection operation and when a sudden change in curvature occurs in the RS curve search result; wherein, when the RS curve search result is a straight-curved curve type, the length of the straight line in the RS curve search result is greater than or equal to the length of the curve.

[0048] A third aspect of an embodiment of the present application provides an electronic device, including a memory and a processor, wherein the memory is used to store a computer program, and the processor runs the computer program to enable the electronic device to execute the parking path generation method described in any one of the first aspects of the embodiment of the present application.

[0049] A fourth aspect of an embodiment of the present application provides a computer-readable storage medium storing computer program instructions. When the computer program instructions are read and executed by a processor, the parking path generation method described in any one of the first aspects of the embodiment of the present application is executed. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments of the present application. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can be obtained based on these drawings without creative work.

[0051] Figure 1 A flowchart of a parking path generation method provided in an embodiment of the present application;

[0052] Figure 2 A flowchart of another parking path generation method provided in an embodiment of the present application;

[0053] Figure 3 A schematic diagram of the structure of a parking path generation device provided in an embodiment of the present application;

[0054] Figure 4 A schematic structural diagram of another parking path generation device provided in an embodiment of the present application;

[0055] Figure 5 A schematic diagram of a dual-resolution grid map provided in an embodiment of the present application;

[0056] Figure 6 A schematic diagram of a hybrid A-star expansion direction provided in an embodiment of the present application. DETAILED DESCRIPTION

[0057] The technical solutions in the embodiments of the present application will be described below in conjunction with the drawings in the embodiments of the present application.

[0058] It should be noted that similar reference numerals and letters represent similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined or explained in subsequent drawings. At the same time, in the description of this application, the terms "first", "second", etc. are only used to distinguish the description and should not be understood as indicating or implying relative importance.

[0059] Example 1

[0060] Please see Figure 1 , Figure 1 A flow chart of a parking path generation method is provided for this embodiment. The parking path generation method includes:

[0061] S101: Determine a parking starting point corresponding to the initial position information of the vehicle in a pre-generated dual-resolution raster map, and add the parking starting point to a preset list of candidate points.

[0062] S102: Pop out the node with the lowest cost from the list of candidate points, and add the node with the lowest cost to the preset list of selected points.

[0063] S103 , performing RS curve search based on the nodes in the selected point list to obtain RS curve search results.

[0064] In this embodiment, the Reeds-Shepp planning algorithm (RS) is a classic geometry-based path planning method with advantages such as simple implementation and fast solution speed.

[0065] S104: Determine whether the RS curve search result passes the obstacle collision detection. If so, execute step S105; if not, end this process.

[0066] S105 : Generate a parking path based on the RS curve search results.

[0067] In this embodiment, the 48 cases of RS curves can be divided into 5 types according to the planning type: CCC, CSC, CCCC, CSCC, and CCSCC, where C represents a curve and S represents a straight line. It can be seen that the RS curves all end with a curve (the distance of the last curve can be very short). For narrow parking conditions, the limitation of parking spaces leads to fewer RS ​​curves that meet the requirements during the car storage stage. In order to make the RS curve play a greater role and improve the success rate of path planning, it is considered to reverse the end point and the starting point, that is, to use the A-star node expansion first in the narrow place of the end point, and to use the RS curve in the more spacious location of the starting point. At this time, both the A-star algorithm and the RS curve algorithm can play their respective advantages, thereby reducing the number of iterations of the hybrid A-star and improving the success rate and running speed of the algorithm. The final planned path can be obtained by simply flipping the obtained planning curve.

[0068] In this embodiment, if the path search and the grid are set to the same resolution, when the grid resolution is set too small, the length of the node extension is too long, which will cause the path search to fail; if the grid resolution is set too large, it will cause the A-star stage to run for too long. In relatively narrow working conditions, the width of the parking space is small, and the longitudinal vertical distance from the parking space to the opposite side is short. At this time, the vehicle needs to perform a certain number of parking behaviors at the entrance of the parking space. If the expansion distance of each node is too long, the expanded node will directly cause the planning to fail due to obstacle collision. Therefore, the node expansion distance should not be too long. To address this problem, a dual-resolution method is adopted to decouple the path resolution and the A-star grid resolution. According to simulation experiments, the A-star grid resolution is set to a (the resolution is inversely proportional to the size, that is, the grid size is 1 / a×1 / a), and the path search grid resolution is set to b (that is, the grid size is 1 / b×1 / b).

[0069] Please see Figure 5 , Figure 5 A schematic diagram showing a dual-resolution raster map.

[0070] In this embodiment, the execution subject of the method may be a computing device such as a computer or a server, and this is not limited in this embodiment.

[0071] In this embodiment, the execution subject of the method may also be a smart device such as a smart phone, a tablet computer, etc., which is not limited in this embodiment.

[0072] It can be seen that the implementation of the parking path generation method described in this embodiment can effectively cope with parking scenarios of various parking spaces in reality by designing a path planning algorithm for narrow parking spaces that are prone to planning failures using the hybrid A-star algorithm. At the same time, the method can also adopt a dual-resolution method to ensure the effectiveness and speed of the algorithm. In addition, the method can also adopt a method of exchanging the end point and the starting point to allow the two hybrid A-star algorithms to play a greater role, thereby making the calculation faster and more real-time. Finally, the method introduces the properties of the clothoid curve to reduce the number of curvature mutations, thereby avoiding parking and turning operations during the parking process and improving the user experience of the parking process.

[0073] Example 2

[0074] Please see Figure 2 , Figure 2 A flow chart of a parking path generation method is provided for this embodiment. The parking path generation method includes:

[0075] S201: Acquire parking space information, vehicle initial position information, and / or obstacle position information.

[0076] In this embodiment, the method can obtain parking space information, vehicle initial position information point coordinates and vehicle angle, and obstacle position (if any) through the front-end sensing module on the vehicle, and process it into a dual-resolution grid map such as Figure 5 As shown in the figure, based on the vehicle parameters and the worst case scenario for obstacle detection in the grid, where the center of the rear axle and the obstacle are located on the grid diagonal, the A-star grid resolution is set to a (denoted as grid 1). On the other hand, the path search grid resolution should be set to b (denoted as grid 2) to minimize computational speed.

[0077] S202 : Generate a dual-resolution grid map based on the parking space information, the vehicle initial position information, and / or the obstacle position information.

[0078] S203: Determine a parking starting point corresponding to the initial position information of the vehicle in the pre-generated dual-resolution grid map.

[0079] S204: Determine a parking destination corresponding to the parking space information in the pre-generated dual-resolution raster map.

[0080] S205: Determine whether there are obstacles at the parking start point and the parking end point. If so, end this process; if not, execute step S206.

[0081] In this embodiment, the method may interchange the parking start point and the parking end point and determine whether they are passable.

[0082] In this embodiment, the starting and ending coordinates of the vehicle are determined based on environmental information and the information is exchanged. Then, it is determined whether there are obstacles at the starting and ending points. If so, the planning fails and the program exits. If there are no obstacles, the information of the planned starting point is added to the OpenList (a list of candidate points, which stores the nodes that can be reached in the current state).

[0083] S206: Add the parking starting point to a preset list of candidate points.

[0084] In this embodiment, the list of points to be selected is called openlist, and the list of selected points is called closelist.

[0085] In this embodiment, the method can pop out the node with the lowest cost from the openlist and add it to the CloseList (a selected point list for storing the judged nodes).

[0086] S207: Pop out the node with the lowest cost from the list of candidate points, and add the node with the lowest cost to the preset list of selected points.

[0087] In this embodiment, the method may pop out the node with the lowest cost from the OpenList and add it to the CloseList. If there is no node in the OpenList, the planning fails.

[0088] S208 , performing RS curve search based on the nodes in the selected point list to obtain RS curve search results.

[0089] S209: When a sudden change in curvature occurs in the RS curve search result, the RS curve search result is smoothed by combining the properties of the clothoid curve.

[0090] In this embodiment, when the RS curve search result is a curve type of straight-curved, the length of the straight line in the RS curve search result is greater than or equal to the length of the curve.

[0091] In this embodiment, the method sets the discrimination condition for the situation where the curvature of the RS curve being searched may suddenly change, combining the properties of the clothoid curve. Each time the curvature changes, a straight line transition is required in the middle or the vehicle's driving direction changes.

[0092] In this embodiment, the method can calculate the length (Sh) of the clothoid curve according to the target vehicle speed (V), the target steering wheel speed (ω), and the target steering wheel angle (θ).

[0093] Sh=(θ / ω)×V

[0094] After obtaining Sh, calculate the curvature change rate (C) based on the minimum turning radius (Rmin):

[0095] C=1 / (Rmin×Sh)

[0096] Finally, bring in the clothoid curve formula to find the coordinates of the end point:

[0097]

[0098]

[0099] According to the experiment, a suitable target steering wheel speed ω = b is selected. At this time, when the length of the clothoid curve Sh = c and the speed V = d, the longitudinal offset y of the clothoid curve relative to the straight line when the curve switches from the minimum turning radius Rmin to -Rmin is H It is about 2cm and can be ignored.

[0100] When using the RS curve for planning, the following judgments need to be made:

[0101] ①Whether this type can generate a path.

[0102] ②If ① is passed, it is necessary to determine whether the direction of the current node is opposite to the first curve of the RS plan. If the directions are opposite (stopping is required when driving in the opposite direction), the curvature is allowed to change suddenly. If they are the same, the curvature must remain consistent.

[0103] ③ When two segments of the RS curve are connected (i.e. CC type), it is necessary to ensure that the directions of the two segments are opposite.

[0104] ④ When the RS curve is of CSC type, it is necessary to determine whether the planned length of the middle straight line segment S is greater than or equal to the length c of the clothoid curve to ensure that the curvature can change smoothly.

[0105] S210 , determining whether the RS curve search result passes the obstacle collision detection. If so, executing step S216 ; if not, executing step S211 .

[0106] In this embodiment, the method performs RS curve search on the nodes popped out from the openlist, and performs obstacle collision detection on the obtained results. If there is no obstacle, a parking path is generated, otherwise the subsequent steps are executed.

[0107] S211. Perform node expansion based on the hybrid A-star algorithm in the dual-resolution grid map to obtain expanded nodes.

[0108] S212: Add the extended node to the list of candidate points.

[0109] S213: Check whether there is a candidate node in the candidate point list that is repeated with the extended node. If so, execute step S214; if not, end this process.

[0110] S214. Calculate the cost of expanding the node.

[0111] S215 . When the cost of the extended node is less than the cost of the candidate node, the cost of the candidate node is updated based on the cost of the extended node, and the candidate node and the extended node are merged, and then step S207 is triggered.

[0112] Please see Figure 6 , Figure 6 A schematic diagram showing the expansion direction of a hybrid A star is shown.

[0113] In this embodiment, the node expansion is consistent with the RS curve, both of which have 6 directions, such as Figure 6 As shown, the previous straight, left, and right turns, the straight, left, and right turns, and the curve radius corresponding to each front wheel steering angle delta is set to the minimum turning radius Rmin. If the current posture is [x, y, θ], the expanded node posture is:

[0114] x=x+k×D2×cos(θ);

[0115] y=y+k×D2×sin(θ);

[0116] θ=θ+k×D2 / L×tan(delta);

[0117] Where D2 is the size of grid 2, and k is the coefficient that ensures that the expanded nodes will not fall in the same grid and produce the phenomenon of "standing still", because the coefficient of the grid diagonal is Therefore, k can be taken as 1.5.

[0118] Each time a node is expanded, the cost of the expanded node to the end point and the node's subscripts in grid 1 and grid 2 must be calculated. The expanded node is compared with the nodes in the openlist based on the corresponding grid 2 subscript and heading angle. If the node already exists, the cost is compared and the smaller cost is updated. If it does not exist, the node is added to the openlist. The cost calculation method for the current node is:

[0119] f=g+h;

[0120] h is the distance from the current node to the end point:

[0121] h=W h ×(cost A× +cost offer );

[0122] Where W h is the weight; cost A× The cost of A star from the current node's grid 1 to the end node's grid 1. Since the cost of A star is calculated based on the grid center, h needs to add the cost from the current position to the center of grid 1, which is defined as cost offer .

[0123] g is the cost from the starting point to the current node, defined as:

[0124] g i =g i-1 +dis(i, i-1)+a×cost change +b×cost turn ;

[0125] Where g i , g i-1 Represent the g cost of the current node and its parent node respectively, dis(i, i-1) represents the distance from the parent node to the current node, cost change Represents the switching cost (forward and backward switching), cost turn represents the cost of steering wheel change, and a and b are coefficients.

[0126] In this embodiment, to ensure that the node extension curvature does not abruptly change, a corresponding determination condition is also introduced in conjunction with the clothoid curve properties described in the above steps. Specifically, the following conditions are determined: whether the current node is in the same direction as the extension. If the directions are opposite, the curvature corresponding to the extension node is not restricted. If they are, the corresponding path of the current node is determined to be a curve. If it is a curve, the extension node can be a curve or a straight line with the same curvature. If it is a straight line, the parent node and the parent-parent node are also straight lines. Only if both are straight lines can the extension node be a curve; otherwise, the extension continues in a straight line.

[0127] S216 : Generate a parking path based on the RS curve search results.

[0128] In this embodiment, when the Euclidean distance deviation between the end point of the curve planned by the RS curve and the expected end point and the vehicle heading angle deviation are both less than the threshold, the planning is successful, the path scatter points are traversed from the closelist, and the scatter points are flipped, and the straight line segments that need to be converted into spiral curves are processed to obtain the final planned path.

[0129] In this embodiment, in order to solve the problem of sudden changes in curve curvature in the path planned by the traditional hybrid A-star algorithm, some restrictions are added to the algorithm's solution process. Every time the curvature changes, a straight line transition is required in the middle. The straight line is combined with the maximum allowable steering wheel speed and then processed into a corresponding clothoid curve, so that the curvature of the planned curve changes continuously.

[0130] In this embodiment, the execution subject of the method may be a computing device such as a computer or a server, and this is not limited in this embodiment.

[0131] In this embodiment, the execution subject of the method may also be a smart device such as a smart phone, a tablet computer, etc., which is not limited in this embodiment.

[0132] It can be seen that the implementation of the parking path generation method described in this embodiment can effectively cope with parking scenarios of various parking spaces in reality by designing a path planning algorithm for narrow parking spaces that are prone to planning failures using the hybrid A-star algorithm. At the same time, the method can also adopt a dual-resolution method to ensure the effectiveness and speed of the algorithm. In addition, the method can also adopt a method of exchanging the end point and the starting point to allow the two hybrid A-star algorithms to play a greater role, thereby making the calculation faster and more real-time. Finally, the method introduces the properties of the clothoid curve to reduce the number of curvature mutations, thereby avoiding parking and turning operations during the parking process and improving the user experience of the parking process.

[0133] Example 3

[0134] Please see Figure 3 , Figure 3 This is a schematic diagram of the structure of a parking path generation device provided in this embodiment. Figure 3 As shown, the parking path generating device includes:

[0135] a determination unit 310 for determining a parking starting point corresponding to the initial position information of the vehicle in a pre-generated dual-resolution raster map, and adding the parking starting point to a preset list of candidate points;

[0136] An adding unit 320 is used to pop out the node with the lowest cost from the list of candidate points and add the node with the lowest cost to a preset list of selected points;

[0137] A search unit 330 is configured to perform an RS curve search based on the nodes in the selected point list to obtain an RS curve search result;

[0138] A judging unit 340 is used to judge whether the RS curve search result passes the obstacle collision detection;

[0139] The generating unit 350 is configured to generate a parking path based on the RS curve search result when the RS curve search result passes the obstacle collision detection.

[0140] In this embodiment, the explanation of the parking path generating device can refer to the description in embodiment 1 or embodiment 2, and will not be further elaborated in this embodiment.

[0141] It can be seen that the implementation of the parking path generation device described in this embodiment can be used to design a path planning algorithm for narrow parking spaces where hybrid A-star is prone to planning failures, thereby effectively coping with parking scenarios in various parking spaces in reality. At the same time, the method can also adopt a dual-resolution method to ensure the effectiveness and speed of the algorithm. In addition, the method can also adopt a method of exchanging the end point and the starting point to allow the two hybrid A-star algorithms to play a greater role, thereby making the calculation speed faster and the real-time performance better. Finally, the method introduces the properties of the clothoid curve, which can reduce the number of curvature mutations, thereby avoiding the operation of parking and turning during the parking process, and improving the user experience of the parking process.

[0142] Example 4

[0143] Please see Figure 4 , Figure 4 This is a schematic diagram of the structure of a parking path generation device provided in this embodiment. Figure 4 As shown, the parking path generating device includes:

[0144] a determination unit 310 for determining a parking starting point corresponding to the initial position information of the vehicle in a pre-generated dual-resolution raster map, and adding the parking starting point to a preset list of candidate points;

[0145] An adding unit 320 is used to pop out the node with the lowest cost from the list of candidate points and add the node with the lowest cost to a preset list of selected points;

[0146] A search unit 330 is configured to perform an RS curve search based on the nodes in the selected point list to obtain an RS curve search result;

[0147] A judging unit 340 is used to judge whether the RS curve search result passes the obstacle collision detection;

[0148] The generating unit 350 is configured to generate a parking path based on the RS curve search result when the RS curve search result passes the obstacle collision detection.

[0149] As an optional implementation manner, the parking path generating device further includes:

[0150] An expansion unit 360 is configured to perform node expansion based on a hybrid A-star algorithm in a dual-resolution grid map to obtain an expanded node when the RS curve search result fails the obstacle collision detection;

[0151] The adding unit 320 is further configured to add the extended node to the list of candidate points, trigger the operation of popping out the node with the lowest cost from the list of candidate points, and adding the node with the lowest cost to the preset list of selected points.

[0152] As an optional implementation manner, the parking path generating device further includes:

[0153] A detection unit 370 is configured to detect whether there is a candidate node in the candidate point list that is repeated with the extended node after the adding unit 320 adds the extended node to the candidate point list;

[0154] A calculation unit 380, configured to calculate the cost of expanding a node when there is a node to be selected;

[0155] The updating unit 390 is configured to update the cost of the node to be selected based on the cost of the extended node when the cost of the extended node is less than the cost of the node to be selected, and merge the node to be selected and the extended node.

[0156] As an optional implementation manner, the parking path generating device further includes:

[0157] The acquiring unit 400 is configured to acquire information about a parking space to be parked, information about the vehicle's initial position, and / or information about an obstacle's position before the determining unit 310 determines the parking starting point corresponding to the vehicle's initial position in the pre-generated dual-resolution raster map.

[0158] The generating unit 350 is further configured to generate a dual-resolution grid map based on the parking space information, the vehicle initial position information and / or the obstacle position information.

[0159] As an optional implementation manner, the parking path generating device further includes:

[0160] The determining unit 310 is further configured to determine a parking destination corresponding to the parking space information to be parked in the pre-generated dual-resolution grid map:

[0161] A determination unit 340 is used to determine whether there are obstacles at the parking start point and the parking end point;

[0162] The determining unit 310 is further configured to add the parking starting point to a preset list of candidate points when there are no obstacles at the parking starting point and the parking end point.

[0163] As an optional implementation manner, the parking path generating device further includes:

[0164] Processing unit 410 is configured to smooth the RS curve search result by combining the properties of a clothoid curve when a sudden change in curvature occurs in the RS curve search result, before the determination unit determines whether the RS curve search result passes the obstacle collision detection operation; wherein, when the RS curve search result is a straight-curved curve type, the length of the straight line in the RS curve search result is greater than or equal to the length of the curve.

[0165] In this embodiment, the explanation of the parking path generating device can refer to the description in embodiment 1 or embodiment 2, and will not be further elaborated in this embodiment.

[0166] It can be seen that the implementation of the parking path generation device described in this embodiment can be used to design a path planning algorithm for narrow parking spaces where hybrid A-star is prone to planning failures, thereby effectively coping with parking scenarios in various parking spaces in reality. At the same time, the method can also adopt a dual-resolution method to ensure the effectiveness and speed of the algorithm. In addition, the method can also adopt a method of exchanging the end point and the starting point to allow the two hybrid A-star algorithms to play a greater role, thereby making the calculation speed faster and the real-time performance better. Finally, the method introduces the properties of the clothoid curve, which can reduce the number of curvature mutations, thereby avoiding the operation of parking and turning during the parking process, and improving the user experience of the parking process.

[0167] An embodiment of the present application provides an electronic device, including a memory and a processor, wherein the memory is used to store a computer program, and the processor runs the computer program to enable the electronic device to execute the parking path generation method in embodiment 1 or embodiment 2 of the present application.

[0168] An embodiment of the present application provides a computer-readable storage medium storing computer program instructions. When the computer program instructions are read and executed by a processor, the parking path generation method in embodiment 1 or embodiment 2 of the present application is executed.

[0169] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can also be implemented in other ways. The device embodiments described above are merely illustrative. For example, the flowcharts and block diagrams in the accompanying drawings show the possible architectures, functions and operations of the devices, methods and computer program products according to the multiple embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of the code, and the module, program segment or a part of the code contains one or more executable instructions for implementing the specified logical functions. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in an order different from that marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart, can be implemented using a dedicated hardware-based system that performs the specified function or action, or can be implemented using a combination of dedicated hardware and computer instructions.

[0170] In addition, the functional modules in each embodiment of the present application can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.

[0171] If the functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0172] The foregoing is merely an embodiment of the present application and is not intended to limit the scope of protection of the present application. Various modifications and variations are possible for those skilled in the art. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included within the scope of protection of the present application. It should be noted that similar reference numerals and letters represent similar items in the following figures. Therefore, once an item is defined in one figure, it does not need to be further defined or explained in subsequent figures.

[0173] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present application should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

[0174] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply the existence of any such actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or device comprising the element.

Claims

1. A parking path generation method, characterized in that: include: Determining a parking starting point corresponding to the initial position information of the vehicle in a pre-generated dual-resolution raster map, and adding the parking starting point to a preset list of candidate points; Popping out the node with the lowest cost from the list of candidate points, and adding the node with the lowest cost to a preset list of selected points; Performing an RS curve search based on the nodes in the selected point list to obtain an RS curve search result; Determine whether the RS curve search result passes the obstacle collision detection; When the RS curve search result passes the obstacle collision detection, generating a parking path based on the RS curve search result; Among them, also include: When the RS curve search result fails the obstacle collision detection, performing node expansion based on a hybrid A-star algorithm in the dual-resolution grid map to obtain an expanded node; Adding the extended node to the list of candidate points, and triggering the step of popping out the node with the lowest cost from the list of candidate points, and adding the node with the lowest cost to the preset list of selected points; Wherein, after the step of adding the extended node to the list of candidate points, the method further includes: Detecting whether there is a candidate node in the candidate point list that is repeated with the extended node; When the candidate node exists, calculating the cost of the extended node; When the cost of the extended node is less than the cost of the node to be selected, the cost of the node to be selected is updated based on the cost of the extended node, and the node to be selected and the extended node are merged.

2. The parking path generation method according to claim 1, characterized in that: Before determining the parking starting point corresponding to the vehicle initial position information in the pre-generated dual-resolution grid map, the method further includes: Obtaining parking space information, vehicle initial position information, and / or obstacle position information; A dual-resolution grid map is generated based on the parking space information, the vehicle initial position information and / or the obstacle position information.

3. The parking path generation method according to claim 1, wherein: The method further comprises: Determine the parking destination corresponding to the parking space to be parked in the pre-generated dual-resolution raster map: Determining whether there are obstacles at the parking starting point and the parking end point; When the obstacles do not exist at the parking starting point and the parking end point, the step of adding the parking starting point to a preset list of candidate points is triggered.

4. The parking path generation method according to claim 1, characterized in that: Before the step of determining whether the RS curve search result passes the obstacle collision detection, the method further includes: When a sudden change in curvature occurs in the RS curve search result, the RS curve search result is smoothed in combination with the properties of the clothoid curve; wherein, when the RS curve search result is a straight-curved curve type, the length of the straight line in the RS curve search result is greater than or equal to the length of the curve.

5. A parking path generation device, characterized in that: The parking path generating device includes: a determining unit, configured to determine a parking starting point corresponding to the initial position information of the vehicle in a pre-generated dual-resolution raster map, and add the parking starting point to a preset list of candidate points; An adding unit, configured to pop out a node with the lowest cost from the list of candidate points, and add the node with the lowest cost to a preset list of selected points; A search unit, configured to perform an RS curve search based on the nodes in the selected point list to obtain an RS curve search result; A judging unit, configured to judge whether the RS curve search result passes the obstacle collision detection; A generating unit, configured to generate a parking path based on the RS curve search result when the RS curve search result passes the obstacle collision detection Among them, also include: An expansion unit, configured to, when the RS curve search result fails the obstacle collision detection, perform node expansion in the dual-resolution grid map based on a hybrid A-star algorithm to obtain an expanded node; The adding unit is further configured to add the extended node to the list of candidate points, and trigger the operation of popping out the node with the lowest cost from the list of candidate points, and adding the node with the lowest cost to the preset list of selected points; Among them, also include: a detection unit, configured to detect whether there is a candidate node in the candidate point list that is repeated with the extension node after the adding unit adds the extension node to the candidate point list; a calculation unit, configured to calculate the cost of the extended node when the candidate node exists; An updating unit is configured to update the cost of the node to be selected based on the cost of the extended node when the cost of the extended node is less than the cost of the node to be selected, and merge the node to be selected and the extended node.

6. An electronic device, characterized in that: The electronic device includes a memory and a processor, the memory is used to store a computer program, and the processor runs the computer program to enable the electronic device to perform the parking path generation method according to any one of claims 1 to 4.

7. A readable storage medium, characterized in that: The readable storage medium stores computer program instructions, and when the computer program instructions are read and executed by a processor, the parking path generation method according to any one of claims 1 to 4 is executed.

Images