Parking path planning method and device, vehicle and electronic equipment

Abstract

The application relates to a parking path planning method and device, a vehicle and electronic equipment, and relates to the technical field of vehicles. The method comprises the following steps: determining a target position point based on obstacle information around a target parking space, wherein the target position point is a projection point of a center point of a rear axle of a vehicle after parking on the target parking space; determining a target planning map, wherein the target planning map comprises but is not limited to a grid map and a quadtree map; the grid map and the quadtree map are both constructed based on the obstacle information; and planning a parking path of the vehicle based on the target position point and the target planning map, so that the robustness of the parking path planning can be improved and different parking environments can be adapted.

Description

Technical Field

[0001] This application relates to the field of vehicle technology, specifically to a parking path planning method, device, vehicle, and electronic equipment. Background Technology

[0002] In an automatic parking system, the path planning module plays a crucial role. Its core task is to accurately plan a collision-free trajectory from the starting position to the ending position based on the surrounding spatial environment of the vehicle.

[0003] Currently, common parking path planning algorithms mainly generate segmented paths based on vehicle kinematic models and parking space geometric parameters. These algorithms have significant advantages such as simple computation and high real-time performance, and can directly generate relatively smooth paths, meeting the basic requirements of automated parking to a certain extent. However, when faced with complex parking scenarios, these algorithms exhibit poor environmental adaptability and are prone to planning failures, thus failing to provide effective path guidance for the vehicle and limiting their application in complex environments. Summary of the Invention

[0004] One of the objectives of this invention is to provide a parking path planning method, apparatus, vehicle, and electronic device that can adapt to different parking environments and improve the robustness of parking path planning.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0006] According to a first aspect of the present invention, a parking path planning method is provided, the method comprising: determining a target location point based on obstacle information surrounding a target parking space, wherein the target location point is the projection point of the center point of the rear axle of the vehicle after parking onto the target parking space; determining a target planning map, wherein the target planning map includes, but is not limited to, a grid map and a quadtree map; both the grid map and the quadtree map are constructed based on obstacle information; and planning a parking path for the vehicle based on the target location point and the target planning map.

[0007] Based on the aforementioned technical means, this application pre-plans target points using obstacle information, avoiding collisions between the parking path and obstacles and ensuring parking positioning accuracy. Furthermore, determining the target point provides clear geometric constraints for subsequent path planning, making it more targeted. In addition, grid maps, quadtree maps, and other maps can be dynamically selected as the target planning map. This dynamic map selection mechanism allows the parking system to cope with parking needs in different scenarios, improving its robustness and generalization ability, and making it suitable for diverse parking environments. Finally, based on the target location point and the dynamically selected target planning map, a collision-free parking path from the starting point to the target point can be generated quickly and accurately.

[0008] In one possible approach, determining the target planning map includes: determining the target planning map based on the environmental complexity of parking, wherein if the environmental complexity is greater than or equal to a preset complexity threshold, the target planning map is a raster map; if the environmental complexity is less than the preset complexity threshold, the target planning map is a quadtree map.

[0009] Based on the aforementioned technical means, this application can dynamically select either a grid map or a quadtree map as the final planning map according to the environmental complexity of parking and a preset complexity threshold, so that the parking system can not only cope with the high-precision parking needs of complex scenarios, but also ensure parking efficiency in simple scenarios, thereby improving the success rate of parking path planning.

[0010] In one possible approach, environmental complexity is determined based on the environmental type that the target parking space conforms to. Environmental types include: narrow aisle, dead-end parking space, and merged parking space; where narrow aisle means that the width of the aisle in front of the parking space is less than the preset width; dead-end parking space means that one of the left or right sides of the parking space is adjacent to a wall; merged parking space means that there are obstacles on both the left and right sides of the parking space.

[0011] In one possible approach, determining the target location point based on obstacle information surrounding the target parking space includes: determining an ideal location point, where the ideal location point is the projection of the center point of the rear axle of the vehicle onto the target parking space under ideal conditions; determining the available parking space for the vehicle based on the parking space boundaries and obstacle information of the target parking space; and determining the target location point based on the ideal location point and the available parking space.

[0012] Based on the aforementioned technical means, this application can filter available parking spaces by combining the parking space boundary and obstacle information of the target parking space, eliminating the influence of obstacles in advance and reducing the probability of vehicles scraping against parking space boundaries and obstacles. Furthermore, by automatically determining the target location point based on the ideal location point and available parking space, it can provide a core positioning target for the automatic parking system, ensuring parking positioning accuracy.

[0013] In one possible approach, the spatial boundary of the parking space is a rectangle; the four sides of the rectangle correspond one-to-one with the four sides of the parking space boundary; based on the ideal position point and the parking space, the target position point is determined, including: if the length and width of the rectangle are greater than the length and width of the vehicle, respectively, calculating the offset distance between the four sides of the rectangle and their corresponding parking space sides. The ideal position point is adjusted based on the offset distance to obtain the target position point.

[0014] One possible approach, based on an ideal location point and available parking space, is to determine the target location point as the ideal location point if the length of the rectangle is less than the length of the vehicle, or the width of the rectangle is less than the width of the vehicle.

[0015] One possible approach involves planning the vehicle's parking path based on the target location and a target planning map. This includes: planning multiple paths on the target planning map using the Hybrid A* algorithm, starting from the vehicle's current rear axle center point and ending at the target location. Then, selecting the optimal path from the multiple paths based on their lengths, the number of reversing segments, and the number of curvature changes. Finally, determining the vehicle's parking path based on the optimal path.

[0016] Based on the aforementioned technical means, this application can utilize the Hybrid A* algorithm, combining vehicle kinematic constraints (such as minimum turning radius and wheelbase) with environmental obstacle information, to quickly generate multiple feasible parking paths using the starting point and target location as anchor points. Furthermore, based on multi-dimensional information such as path length, number of reversing segments, and number of curvature changes, it can ensure that the selected optimal path is both efficient and safe.

[0017] In one possible approach, on the target planning map, the Hybrid A* algorithm is used to plan multiple paths starting from the vehicle's current rear axle center point and ending at the target location point. This includes: during each node expansion process, generating multiple feasible motion trajectories starting from the first target node based on the vehicle's maximum permissible steering angle. The intersection of these multiple motion trajectories with the first grid line of the target planning map is determined as a first candidate node. The first candidate node that meets preset conditions is determined as a second candidate node, where the preset conditions constrain the vehicle to have no collision risk with obstacles during its passage through the first candidate node; the collision risk is determined based on a distance map. Based on the Venn diagram, the path cost of the second candidate node is calculated, where the Venn diagram and distance map are determined based on obstacle information. The second candidate node with the minimum path cost is determined as the second target node. Based on the second target node, a path from the starting point to the ending point is generated.

[0018] In the initial node expansion process, the first target node is the starting point; if the second target node is the ending point, the node expansion ends; if the second target node is not the ending point, the second target node becomes the first target node in the next node expansion process.

[0019] Based on the aforementioned technical methods, this application incorporates the maximum permissible steering angle of the vehicle into the node expansion process, ensuring that all motion trajectories are executable paths for the vehicle. Furthermore, by combining Venn diagram quantification with the distribution of obstacles around the nodes, the path cost includes not only distance cost but also implicitly the proximity to obstacles. This approach makes the algorithm inclined to select the path with the least obstacle influence, improving the safety of parking paths in dynamic scenarios.

[0020] In one possible approach, the path cost of the second candidate node is calculated based on the Venn diagram, including: estimating the path cost between the second candidate node and the endpoint based on the Venn diagram; determining the path cost between the first target node and the second candidate node based on changes in travel direction, steering wheel angle, and gear shifting between the first target node and the second candidate node; and determining the path cost of the second candidate node as the sum of the cumulative path cost from the starting point to the first target node, the path cost between the first target node and the second candidate node, and the path cost between the second candidate node and the endpoint.

[0021] In one possible approach, based on a Venn diagram, the path cost between the second candidate node and the endpoint is estimated, including: determining the maximum and minimum distances between the second candidate node and the Venn edges in the Venn diagram; calculating a first cost based on the maximum and minimum distances, where a smaller first cost indicates that the second candidate node is closer to a Venn edge; calculating a second cost based on the distance between the second candidate node and obstacles; determining a third cost based on the distance between the second candidate node and the endpoint; determining a fourth cost based on the distance between the second candidate node and the starting point; determining a fifth cost based on the distance between the second candidate node and the guide point of the target parking space; and estimating the path cost between the second candidate node and the endpoint based on the first, second, third, fourth, and fifth costs.

[0022] In one possible approach, based on the second target node, a path from the starting point to the ending point is generated, including: using a Reeds-Shepp (RS) curve planner and / or a geometric path planner to generate a planned path from the second target node to the ending point; concatenating the extended paths between the first and second target nodes during the previous node expansion process to obtain a first path; and concatenating the first path with the planned path to obtain the path from the starting point to the ending point.

[0023] Based on the aforementioned technical means, this application can generate parking paths that balance smoothness, obstacle avoidance capability, and high efficiency by using an RS curve planner or geometric path planner, combined with node expansion and path splicing strategies, thereby improving the quality of parking paths.

[0024] In one possible approach, the geometric path planner uses a two-stage path planning algorithm to generate a planned path from the second target node to the termination point. The two-stage path planning algorithm aims to minimize the sum of the first arc from the second target node to the first intermediate node and the second arc from the first intermediate node to the termination point, and optimizes the solution by ensuring that the tangent vectors of the circles containing the first and second arcs are continuous at the first intermediate node.

[0025] In one possible approach, the geometric path planner uses a five-segment path planning algorithm to generate the planned path from the second target node to the endpoint. The five-segment path planning algorithm optimizes the solution by minimizing the straight line segment between the second target node and the second intermediate node, the arc segment between the second intermediate node and the third intermediate node, the arc segment between the third intermediate node and the fourth intermediate node, the arc segment between the fourth intermediate node and the fifth intermediate node, and the distance between the fifth intermediate node and the endpoint.

[0026] The position of the third intermediate node and the radius of the circles containing the arc segments from the second and third intermediate nodes, the third and fourth intermediate nodes, and the fourth and fifth intermediate nodes are predetermined.

[0027] One possible approach involves determining the vehicle's parking path based on the optimal path, including: generating a static target corridor (STC) region based on a quadtree map, where the STC region is a T-shaped area enclosed by the target parking space and the obstacle-free area defined by the quadtree map; determining the ideal path point corresponding to each path point on the optimal path within the STC region; ensuring the ideal path point is close to the centerline of the STC region; and aiming to smooth the optimal path by minimizing the sum of the differences between each path point on the smoothed optimal path and its corresponding ideal path point, as well as minimizing the smoothing loss of adjacent path points on the smoothed optimal path. Finally, determining the vehicle's parking path based on the smoothed optimal path.

[0028] Based on the aforementioned technical means, this application uses the center line of the STC area as a reference for the ideal path, ensuring that the path points are close to the center of the channel, avoiding being too close to obstacles, reducing the risk of collision, and improving path safety.

[0029] One possible approach involves determining the vehicle's parking path based on the smoothed optimal path, including: finding target path point pairs on the smoothed optimal path whose distance is less than a preset distance threshold and whose directional angle difference is less than a preset angle; trimming the path between the target path point pairs if the steering angle between them is less than a preset steering angle; and determining the trimmed optimal path as the vehicle's parking path.

[0030] Based on the aforementioned technical means, this application can, on the basis of the smoothed optimal path, filter target path point pairs through triple constraints of distance threshold, direction angle difference, and steering angle, and trim redundant paths between two points to generate a simpler parking path. This method improves parking efficiency and smoothness by removing invalid curves or detours in the path.

[0031] According to a second aspect of the present invention, a parking route planning device is provided, the device comprising: a first determining unit, a second determining unit, and a planning unit.

[0032] The first determining unit is used to determine the target location point based on obstacle information around the target parking space; wherein, the target location point is the projection point of the center point of the rear axle of the vehicle after parking onto the target parking space.

[0033] The second determining unit is used to determine the target planning map, wherein the target planning map includes, but is not limited to, a grid map and a quadtree map; both the grid map and the quadtree map are constructed based on obstacle information.

[0034] The planning unit is used to plan the parking path of the vehicle based on the target location point and the target planning map.

[0035] In one possible approach, the second determining unit is specifically used to determine a target planning map based on the environmental complexity of parking, wherein if the environmental complexity is greater than or equal to a preset complexity threshold, the target planning map is a raster map; if the environmental complexity is less than the preset complexity threshold, the target planning map is a quadtree map.

[0036] In one possible approach, the first determining unit is specifically used to: determine an ideal location point, wherein the ideal location point is, under ideal conditions, the projection of the center point of the rear axle of the vehicle onto the target parking space; determine the available parking space for the vehicle based on the parking space boundaries and obstacle information of the target parking space; and determine the target location point based on the ideal location point and the available parking space.

[0037] In one possible approach, the first determining unit further includes a calculation subunit and a processing subunit. The calculation subunit calculates the offset distances between the four sides of the rectangle and their corresponding parking space sides, provided that the length and width of the rectangle are greater than the length and width of the vehicle, respectively. The processing subunit adjusts the ideal position point based on the offset distances to obtain the target position point.

[0038] In one possible approach, the first determining unit is specifically used to determine the target location point as the ideal location point when the length of the rectangle is less than the length of the vehicle, or the width of the rectangle is less than the width of the vehicle.

[0039] In one possible approach, the planning unit further includes a planning sub-unit, a selection sub-unit, and a determination sub-unit. The planning sub-unit is used on the target planning map to plan multiple paths starting from the vehicle's current rear axle center point and ending at the target location point using the Hybrid A* algorithm. The selection sub-unit is used to select the optimal path from the multiple paths based on the path lengths, the number of reversing segments, and the number of curvature changes. The determination sub-unit is used to determine the vehicle's parking path based on the optimal path.

[0040] In one possible approach, the planning sub-unit is specifically used to generate multiple feasible motion trajectories starting from the first target node, based on the vehicle's maximum permissible steering angle, during each node expansion process. The intersection of these multiple motion trajectories with the first grid line of the target planning map is determined as a first candidate node. The first candidate node that meets preset conditions is determined as a second candidate node, where the preset conditions constrain the vehicle to have no collision risk with obstacles during its passage through the first candidate node; the collision risk is determined based on a distance map. The path cost of the second candidate node is calculated based on a Venn diagram, where the Venn diagram and distance map are determined based on obstacle information. The second candidate node with the minimum path cost is determined as the second target node. Based on the second target node, a path from the starting point to the ending point is generated.

[0041] In the initial node expansion process, the first target node is the starting point; if the second target node is the ending point, the node expansion ends; if the second target node is not the ending point, the second target node becomes the first target node in the next node expansion process.

[0042] In one possible approach, the sub-unit is specifically used to generate a STC region based on a quadtree map, where the STC region is a T-shaped area enclosed by the target parking space and the obstacle-free area defined by the quadtree map. The ideal path point corresponding to each path point on the optimal path is determined within the STC region; the ideal path point is close to the centerline of the STC region. The goal of smoothing the optimal path is to minimize the sum of the differences between each path point on the smoothed optimal path and its corresponding ideal path point, and to minimize the smoothing loss of adjacent path points on the smoothed optimal path. Based on the smoothed optimal path, the parking path for the vehicle is determined.

[0043] According to a third aspect of the present invention, an electronic device is provided, comprising: a processor; a memory for storing processor-executable instructions; wherein the processor is configured to execute instructions to implement the method of the first aspect described above and any possible implementation thereof.

[0044] According to a fourth aspect of the present invention, a computer-readable storage medium is provided, which, when the instructions in the computer-readable storage medium are executed by a processor of a processing device, enables the processing device to perform the methods described in the first aspect and any possible embodiments thereof.

[0045] According to a fifth aspect of the present invention, a computer program product is provided, the computer program product including computer instructions that, when executed on a processing device, cause the processing device to perform the method described in the first aspect and any possible implementation thereof.

[0046] Therefore, the above-mentioned technical features of the present invention have the following beneficial effects:

[0047] (1) By planning target points in advance using obstacle information, collisions between the parking path and obstacles can be avoided, and parking positioning accuracy can be ensured. At the same time, the determination of target points can provide clear geometric constraints for subsequent path planning, making the path planning more targeted. In addition, based on the complexity of the parking environment and the preset complexity threshold, the system dynamically selects between a grid map and a quadtree map. This mechanism of dynamically switching map types enables the parking system to cope with the high-precision requirements of complex scenarios while maintaining high efficiency in simple scenarios, improving the robustness and generalization ability of the parking system, and making it suitable for diverse parking environments. Based on the target location point and combined with the dynamically selected target planning map, a collision-free parking path from the starting point to the target point can be generated quickly and accurately.

[0048] (2) Parking spaces can be screened by combining the parking space boundary and obstacle information of the target parking space, eliminating the impact of obstacles in advance and reducing the probability of vehicles scraping against parking space boundaries and obstacles. In addition, the target location point can be automatically determined based on the ideal location point and the parking space, which can provide the core positioning target for the automatic parking system and ensure parking positioning accuracy.

[0049] (3) The Hybrid A* algorithm can be used to combine vehicle kinematic constraints (such as minimum turning radius and wheelbase) with environmental obstacle information to quickly generate multiple feasible parking paths with the starting point and target location as anchor points. In addition, based on multi-dimensional information such as path length, number of reversing segments and number of curvature changes, the selected optimal path is ensured to be efficient and safe.

[0050] (4) The maximum permissible steering angle of the vehicle is incorporated into the node expansion to ensure that all motion trajectories are executable paths for the vehicle. In addition, by combining the distribution of obstacles around the node with the Venn diagram quantification, the path cost not only includes distance cost but also implicitly includes the proximity to obstacles. This approach makes the algorithm tend to select the path with the least impact from obstacles, thereby improving the safety of parking paths in dynamic scenarios.

[0051] (5) Use the center line of the STC area as a reference for the ideal path to ensure that the path points are close to the center of the channel, avoid getting too close to obstacles, reduce the risk of collision, and improve the safety of the path.

[0052] (6) Based on the smoothed optimal path, target path point pairs can be filtered through triple constraints of distance threshold, direction angle difference, and steering angle to trim redundant paths between two points and generate a simpler parking path. This method improves parking efficiency and smoothness by removing invalid curves or detours in the path. Attached Figure Description

[0053] Figure 1 A schematic diagram illustrating the implementation environment of a parking path planning method provided in this application embodiment;

[0054] Figure 2 A flowchart illustrating a parking path planning method provided in an embodiment of this application;

[0055] Figure 3 A flowchart illustrating another parking path planning method provided in an embodiment of this application;

[0056] Figure 4 A schematic diagram of the structure of a parking path planning device provided in an embodiment of this application;

[0057] Figure 5 This is a block diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0058] To enable those skilled in the art to better understand the technical solutions of this application, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings.

[0059] It should be noted that the terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0060] In the embodiments of this application, the words "exemplary," "for example," or "e.g.," are used to indicate examples, illustrations, or explanations. Any embodiment or design described as "exemplary," "for example," or "e.g.," in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the words "exemplary," "for example," or "e.g.," is intended to present the relevant concepts in a specific manner.

[0061] First, the relevant technologies involved in this application will be explained to facilitate understanding by those skilled in the art.

[0062] With the rapid development of autonomous driving technology, automated parking, as one of its key application scenarios, has received widespread attention. Path planning, as the core module of automated parking, can plan a collision-free trajectory from the starting position to the ending position based on the spatial environment surrounding the vehicle. This trajectory must not only strictly conform to the vehicle's dynamic characteristics to ensure that the vehicle can stably and accurately follow the trajectory, but also fully consider human driving habits to improve the comfort of the automated parking function and bring a better user experience.

[0063] Currently, common parking path planning algorithms mainly consist of geometry / motion model-based algorithms and sampling search-based algorithms. Among them, geometry / motion model-based algorithms generate segmented paths based on the vehicle's kinematics model and the geometric parameters of the parking space. This algorithm has significant advantages such as simple computation and high real-time performance, and can directly generate relatively smooth paths, meeting the basic requirements of automatic parking to a certain extent. However, this algorithm also has obvious limitations. When faced with complex and changing parking scenarios, this algorithm has poor environmental adaptability and is prone to planning failures, thus failing to provide effective path guidance for the vehicle and limiting its application in complex environments.

[0064] Sampling search-based planning algorithms have the advantage of generating feasible paths with direction and curvature continuity. This allows vehicles to travel more smoothly and steadily along the planned path, resulting in relatively high path quality. However, they also have significant drawbacks. Because they require searching and calculating a large number of sampling points, the algorithm consumes high computational resources, and the generated paths are often not smooth enough, requiring additional smoothing post-processing. This undoubtedly increases the algorithm's complexity and implementation difficulty.

[0065] To address the aforementioned technical problems, this application provides a parking path planning method. By pre-planning target points using obstacle information, collisions between the parking path and obstacles can be avoided, ensuring parking positioning accuracy. Furthermore, determining the target point provides clear geometric constraints for subsequent path planning, making it more targeted. In addition, grid maps, quadtree maps, and other maps can be dynamically selected as the target planning map. This dynamic map type selection mechanism allows the parking system to handle parking needs in different scenarios, improving its robustness and generalization ability, and making it suitable for diverse parking environments. Finally, based on the target location point and the dynamically selected target planning map, a collision-free parking path from the starting point to the target point can be generated quickly and accurately.

[0066] The technical solutions of the embodiments of this application will be described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments.

[0067] The parking path planning method provided in this application can be applied to vehicles. Vehicles can also be referred to as vehicles, mobile carriers, electric vehicles (EVs), hybrid electric vehicles (HEVs), plug-in hybrid electric vehicles (PHEVs), fuel cell vehicles (FCVs), autonomous vehicles, intelligent and connected vehicles (ICVs), driverless vehicles, etc.

[0068] In this application, the vehicle can be a sedan, a sport utility vehicle (SUV), a truck, an electric vehicle, a motorcycle, a tricycle, a special vehicle (such as an ambulance, fire truck, police car, etc.), a driverless taxi, an intelligent connected bus, an autonomous logistics vehicle, an electric truck, etc. Furthermore, this method is also applicable to various special-purpose vehicles, such as agricultural vehicles, mining vehicles, forestry vehicles, airport vehicles, and port vehicles. This application does not impose specific limitations in this regard.

[0069] like Figure 1 As shown, the implementation environment of the parking path planning method provided in this application embodiment includes a parking path planning device 101. The parking path planning device 101 is deployed in the vehicle 100.

[0070] In this embodiment, the parking path planning device 101 can acquire obstacle information around the target parking space and determine a target location point based on the obstacle information. The target location point is the projection of the center point of the rear axle of the vehicle after parking onto the target parking space. Then, the parking path planning device 101 can plan the parking path of the vehicle 100 based on the determined target planning map. Finally, the parking path planning device 101 can plan the parking path of the vehicle 100 based on the target location point and the target planning map.

[0071] Optionally, Figure 1 The parking path planning device 101 can be a terminal, a server, or other types of electronic devices. Figure 1 The diagram shown is merely an example of the device configuration of the parking path planning device 101 and does not constitute a limitation thereof.

[0072] When the parking path planning device 101 is a terminal, the terminal can be a device providing voice and / or data connectivity to a user, a handheld device with wireless connectivity, or other processing devices connected to a wireless modem. The terminal can communicate with one or more core networks via a radio access network (RAN). The terminal can be a mobile terminal, such as a computer with a mobile terminal that exchanges voice and / or data with the radio access network, for example, a mobile phone, tablet computer, laptop computer, netbook, or personal digital assistant (PDA). This application does not impose any limitations on this.

[0073] When the parking path planning device 101 is a server, the server can be a single server or a server cluster consisting of multiple servers. In some embodiments, the server cluster can also be a distributed cluster. This application does not impose any limitations in this regard.

[0074] It should be noted that the structure illustrated in the embodiments of this application does not constitute a limitation on vehicle 100. It may include more or fewer components than illustrated, or combine some components, or split some components, or have different component arrangements. The illustrated components may be implemented in hardware, software, or a combination of software and hardware.

[0075] For ease of understanding, the parking path planning method provided in this application will be described in detail below with reference to the accompanying drawings.

[0076] Figure 2 This is a flowchart illustrating a parking path planning method provided in an embodiment of this application, as shown below. Figure 2 As shown, the method includes:

[0077] S201. Determine the target location point based on the obstacle information around the target parking space.

[0078] The target location point is the projection of the center point of the rear axle of the vehicle after parking onto the target parking space.

[0079] Optionally, the target parking space may include parallel parking spaces, perpendicular parking spaces, and angled parking spaces, which are not limited in this application.

[0080] In some embodiments, the vehicle may first determine an ideal location point, which is the projection of the center point of the vehicle's rear axle onto the target parking space under ideal conditions. Then, the vehicle may determine its available parking space based on the parking space boundaries and obstacle information of the target parking space. Finally, the vehicle may determine the target location point based on the ideal location point and the available parking space.

[0081] In one example, the vehicle can establish a parking coordinate system based on the target parking space and its own direction of entry. Then, based on the target parking space, the vehicle can establish a first initial axis (hereinafter referred to as baseline_x) parallel to the x-axis of the parking coordinate system and a second initial axis (hereinafter referred to as baseline_y) parallel to the y-axis of the parking coordinate system, and use the intersection of baseline_x and baseline_y as the ideal position point. Next, the vehicle can determine the available parking space based on the boundary of the target parking space and the distances between the boundaries of each parking space and obstacles. The boundary of the available parking space is a rectangle, with each of its four sides corresponding to one of the four sides of the parking space boundary.

[0082] Based on the above, after determining the available parking space for the vehicle, if the length and width of the rectangle are greater than the length and width of the vehicle, the vehicle can calculate the offset distance between the four sides of the rectangle and their corresponding parking space sides. Then, the vehicle can adjust the ideal position point based on the offset distance to obtain the target position point. Alternatively, if the length of the rectangle is less than the length of the vehicle, or the width of the rectangle is less than the width of the vehicle, the vehicle can determine the target position point as the ideal position point.

[0083] In this parking coordinate system, the origin is the second corner point of the target parking space after the vehicle passes through it. The x-axis of the parking coordinate system is parallel to the first boundary line of the target parking space, and the positive direction of the x-axis is the direction from the first corner point of the target parking space to the second corner point. The first boundary line is the boundary line between the first and second corner points of the target parking space. The y-axis of the parking coordinate system is determined by the right-hand rule. For example, extend your right hand, keep your four fingers together, and perpendicular to your thumb. If your thumb points in the positive x-axis direction, the direction your four fingers ultimately point is the positive y-axis direction.

[0084] S202. Determine the target and plan the map.

[0085] The target planning map may include, but is not limited to, quadtree maps and grid maps. Both grid maps and quadtree maps are constructed based on obstacle information. The specific construction methods can be found in the following embodiments, and will not be elaborated here.

[0086] In some embodiments, the vehicle can determine the environmental complexity of parking based on the environmental type of the target parking space. Then, if the environmental complexity is greater than or equal to a preset complexity threshold, the vehicle can use a grid map as the target planning map. If the environmental complexity is less than the preset complexity threshold, the vehicle can use a quadtree map as the target planning map.

[0087] In this application's embodiments, the environment type may include narrow passages, dead-end parking spaces, and merged parking spaces, etc., and this application does not limit this. A narrow passage refers to a passageway in front of the parking space whose width is less than a preset width; a dead-end parking space refers to a parking space where one of its left or right sides is adjacent to a wall; and a merged parking space refers to a parking space where there are obstacles on both sides. For example, in environment types with high complexity such as narrow passages, dead-end parking spaces, and merged parking spaces, a grid map can be used as the target planning map; in environment types with low complexity such as parking spaces with obstacles on only one side, a quadtree map can be used as the target planning map.

[0088] S203. Based on the target location point and the target planning map, plan the parking path for the vehicle.

[0089] In some embodiments, the vehicle can plan multiple paths on a target planning map using the Hybrid A* algorithm, starting from the vehicle's current rear axle center point and ending at the target location. Then, the vehicle can select the optimal path from the multiple paths based on the path lengths, the number of reversing segments, and the number of curvature changes. Finally, the vehicle can determine its parking path based on the optimal path.

[0090] The method for planning multiple paths using the Hybrid A* algorithm can be found in the following embodiments, and the method for determining the parking path of a vehicle based on the optimal path can also be found in the following embodiments, and will not be elaborated here.

[0091] For example, for each of multiple paths, the vehicle can determine the cost of that path based on its length, the number of reversing segments, and the number of curvature changes, and select the path with the lowest cost as the optimal path. The formula for calculating the cost of a path can be found in Formula 1 below.

[0092] (Formula 1).

[0093] in, Indicates the first Path The cost, Representing a path Length, Representing a path Number of reversing sections Representing a path The number of curvature changes, Weighting coefficients representing path length. The weighting coefficient represents the number of reversing segments. The weighting coefficient represents the number of curvature changes.

[0094] Based on the above technical solution, target points can be planned in advance using obstacle information to avoid collisions between the parking path and obstacles, ensuring parking positioning accuracy. Furthermore, determining the target points provides clear geometric constraints for subsequent path planning, making it more targeted. In addition, based on the complexity of the parking environment and a preset complexity threshold, a grid map or a quadtree map is dynamically selected. This dynamic map type switching mechanism allows the parking system to handle the high-precision requirements of complex scenarios while maintaining efficiency in simple scenarios, improving the robustness and generalization ability of the parking system, making it suitable for diverse parking environments. Based on the target location point and combined with the dynamically selected target planning map, a collision-free parking path from the starting point to the target point can be generated quickly and accurately, improving the quality of path generation.

[0095] In one alternative implementation, such as Figure 3 As shown, the Hybrid A* algorithm is used to plan multiple paths on the target planning map, starting from the current rear axle center point of the vehicle and ending at the target location point. Specifically, these paths can include:

[0096] S301. During each node expansion process, based on the vehicle's maximum permissible steering angle, generate multiple feasible motion trajectories starting from the first target node.

[0097] The first target node in the initial node expansion process is the starting point.

[0098] Optional, feasible movement trajectories may include forward, backward, left turn, or right turn directions, which can be determined according to the actual situation.

[0099] S302. The intersection point of multiple motion trajectories with the first grid line of the target planning map is determined as the first candidate node.

[0100] S303. The first candidate node that meets the preset conditions is determined as the second candidate node.

[0101] The preset conditions are used to ensure that the vehicle has no risk of collision with obstacles while passing through the first candidate node. The collision risk is determined based on a distance map, which is constructed based on obstacle information. The specific construction method can be found in the following embodiments and will not be elaborated here.

[0102] In some embodiments, the distance map stores the minimum distance between each node (i.e., grid) and an obstacle. Based on this, a vehicle can determine the minimum distance between a first candidate node and an obstacle based on the distance map. If the minimum distance is greater than a distance threshold, it indicates that there is no risk of collision with the obstacle while the vehicle passes through the first candidate node, and the first candidate node is designated as a second candidate node and added to the open list. If the minimum distance between the first candidate node and the obstacle is less than the distance threshold, it indicates that there is a risk of collision with the obstacle while the vehicle passes through the first candidate node, and the first candidate node is added to the closed list.

[0103] S304. Based on the Venn diagram, calculate the path cost of the second candidate node.

[0104] The Veno map is determined based on obstacle information. The specific construction method can be referred to the following embodiments, which will not be repeated here.

[0105] In some embodiments, the vehicle can estimate the path cost between the second candidate node and the endpoint based on a Venn diagram. Then, the vehicle can determine the path cost between the first target node and the second candidate node based on changes in travel direction, steering wheel angle, and gear shifts between them. Finally, the vehicle can determine the path cost of the second candidate node as the sum of the cumulative path cost from the starting point to the first target node, the path cost between the first target node and the second candidate node, and the path cost between the second candidate node and the endpoint.

[0106] For example, the path cost of the second candidate node satisfies the following formula 2, and the path cost between the first target node and the second candidate node satisfies the following formulas 3 and 4.

[0107] (Formula 2) .

[0108] in, This represents the path cost of the second candidate node. This represents the cumulative path cost from the starting point to the first target node. This represents the path cost between the first target node and the second candidate node. This represents the path cost between the second candidate node and the termination point.

[0109] (Formula 3).

[0110] (Formula 4).

[0111] in, This represents the change in direction of the vehicle from the first target node to the second candidate node. This represents the change in steering wheel angle of the vehicle from the first target node to the second candidate node. Indicates the number of gear shifts. Indicates whether to shift gears (e.g., 1 indicates shifting gears, 0 indicates not shifting gears). Indicates whether it is a short path shift (e.g., 1 indicates a short path shift, 0 indicates no short path shift). Weighting coefficients representing the amount of change in direction. The weighting coefficient represents the change in steering wheel angle. The weighting coefficient represents the number of gear shifts. This represents the weighting coefficients of the shift function. This represents the shift function. This indicates the current cumulative number of gear shifts, that is, the cumulative number of gear shifts between the starting point and the second candidate node.

[0112] In one example, the aforementioned estimation of the path cost between the second candidate node and the termination point can be implemented as follows: The vehicle can determine the maximum and minimum distances between the second candidate node and the Vino edge in the Vino graph, and calculate a first cost based on the maximum and minimum distances, where a smaller first cost indicates that the second candidate node is closer to the Vino edge. Then, the vehicle can calculate a second cost based on the distance between the second candidate node and the obstacle. Next, the vehicle can determine a third cost based on the distance between the second candidate node and the termination point. The vehicle can also determine a fourth cost based on the distance between the second candidate node and the starting point, and a fifth cost based on the distance between the second candidate node and the guide point of the target parking space. Finally, the vehicle can estimate the path cost between the second candidate node and the termination point based on the first, second, third, fourth, and fifth costs.

[0113] For example, the path cost between the second candidate node and the termination point satisfies the following formulas 5-7.

[0114]

[0115] (Formula 6).

[0116] (Formula 7).

[0117] in, This represents the distance between the second candidate node and the termination point (such as Euclidean distance or Dubins distance), i.e., the third cost. An indicator function that indicates whether to enable the guide point cost. This indicates the distance between the second candidate node and the guide point of the target parking space. This represents the fifth cost. Indicates the obstacle distance weight. This represents the distance between the second candidate node and the obstacle. This indicates the second cost. This indicates the shift penalty factor. This indicates the number of direction changes for the second candidate node. This represents the threshold for the number of direction changes. This represents the penalty factor from the starting point. Indicates the second candidate node and the starting point distance, Indicates the distance penalty threshold. This indicates the fourth cost.

[0118] This represents the cost of the Vinyson graph, also known as the first cost. This represents the coefficient that adjusts the reward level of the Vinoclot diagram. The weight function represents the cost of the Vinograph. This represents the minimum distance between the second candidate node and the Vino edge in the Vino graph. This represents the maximum distance between the second candidate node and the Vino edge in the Vino graph. This indicates the sensitivity to controlling rewards. Represents the safe distance mapping function, It can be a Gaussian function; there are no restrictions on this.

[0119] S305. The second candidate node with the lowest path cost is determined as the second target node.

[0120] In this embodiment, the first target node in the initial node expansion process is the starting point. If the second target node is the ending point, the node expansion ends; if the second target node is not the ending point, the second target node becomes the first target node in the next node expansion process.

[0121] S306. Based on the second target node, generate the path from the starting point to the ending point.

[0122] In some embodiments, the vehicle can use an RS curve planner and / or a geometry path planner to generate a planned path from the second target node to the end point. Then, the vehicle can concatenate the extended paths between the first and second target nodes during the previous node expansion process to obtain a first path, and concatenate the first path with the planned path to obtain the path from the start point to the end point.

[0123] In this embodiment, the path generated using the RS curve planner is described. It can be composed of five basic segments, namely .in, An arc representing the minimum turning radius of a vehicle. Represents a straight line. It indicates a change of direction, that is, switching from forward to backward, or from backward to forward.

[0124] In this embodiment, the geometric path planner can use a two-segment path planning algorithm to generate the planned path from the second target node to the end point, or it can use a five-segment path planning algorithm to generate the planned path from the second target node to the end point. The following provides a detailed description of a) the geometric path planner using a two-segment path planning algorithm to generate the planned path from the second target node to the end point, and b) the geometric path planner using a five-segment path planning algorithm to generate the planned path from the second target node to the end point.

[0125] a. The geometric path planner can use a two-stage path planning algorithm to generate a planned path from the second target node to the end point.

[0126] The two-stage path planning algorithm aims to minimize the sum of the first arc from the second target node to the first intermediate node and the second arc from the first intermediate node to the termination point (hereinafter referred to as the first constraint condition), and optimizes the solution by ensuring that the tangent vectors of the circle containing the first arc and the circle containing the second arc are continuous at the first intermediate node (hereinafter referred to as the second constraint condition).

[0127] For example, the center of the first arc It satisfies the following formula 8.

[0128] (Formula 8).

[0129] in, Indicates the second target node. This represents the radius of the first arc. This represents the angle between the vehicle's orientation at the second target node (i.e., the direction from the center point of the rear axle to the center point of the front axle) and the x-axis of the parking coordinate system.

[0130] The center of the second arc It satisfies the following formula 9.

[0131] (Formula 9).

[0132] in, Indicates the endpoint. This represents the radius of the second arc. This represents the angle between the vehicle's orientation at the termination point and the x-axis of the parking coordinate system.

[0133] First intermediate node The following formula 10 is satisfied, the first constraint condition is satisfied with the following formula 11, and the second constraint condition is satisfied with the following formula 12.

[0134] (Formula 10).

[0135] (Formula 11).

[0136] (Formula 12).

[0137] in, Indicates the first intermediate node. This represents the sum of the first and second arcs. express The difference between the vehicle's orientation at the first intermediate node and the angle between the vehicle's orientation and the x-axis of the parking coordinate system. This indicates the angle between the vehicle's orientation at the first intermediate node and the x-axis of the parking coordinate system. The difference.

[0138] Based on the above, a system of nonlinear equations is constructed using Equations 8-12, and optimized using the first and second constraints to obtain the planned path from the second target node to the termination point.

[0139] b. The geometric path planner can use a five-segment path planning algorithm to generate a planned path from the second target node to the end point.

[0140] The five-segment path planning algorithm aims to minimize the following distances: the straight line segment between the second target node and the second intermediate node, the arc segment between the second intermediate node and the third intermediate node, the arc segment between the third intermediate node and the fourth intermediate node, the arc segment between the fourth intermediate node and the fifth intermediate node, and the distance between the fifth intermediate node and the termination point (hereinafter referred to as the five-segment distance).

[0141] The position of the third intermediate node and the radii of the arc segments from the second and third intermediate nodes, the third and fourth intermediate nodes, and the fourth and fifth intermediate nodes are predetermined.

[0142] Optionally, the third intermediate node can be located 30cm directly above the corner of the first parking space after the vehicle passes the target parking space; there is no limitation on this.

[0143] For example, the second intermediate node It satisfies the following formula 13.

[0144] (Formula 13).

[0145] in, Indicates along Direction from the second target node To the second intermediate node The distance between the straight line segments.

[0146] Second intermediate node To the third intermediate node The center of the arc segment It satisfies the following formula 14.

[0147] (Formula 14).

[0148] in, Indicates the second intermediate node To the third intermediate node The radius of the arc segment, Indicates that the vehicle is in The angle between the orientation of the vehicle and the x-axis of the parking coordinate system.

[0149] Third intermediate node To the fourth intermediate node The center of the arc segment It satisfies the following formula 15.

[0150] (Equation 15).

[0151] in, Indicates the third intermediate node To the fourth intermediate node The radius of the arc segment, Indicates that the vehicle is in The angle between the orientation of the vehicle and the x-axis of the parking coordinate system.

[0152] Fourth intermediate node To the fifth intermediate node The center of the arc segment It satisfies the following formula 16.

[0153] (Equation 16).

[0154] in, Indicates the fourth intermediate node To the fifth intermediate node The radius of the arc segment, Indicates that the vehicle is in The angle between the orientation of the vehicle and the x-axis of the parking coordinate system.

[0155] End point It satisfies the following formula 17.

[0156] (Equation 17).

[0157] in, Indicates along Direction from the fifth intermediate node To the end point The distance between the straight line segments. Indicates the vehicle is at the termination point. The angle between the orientation of the vehicle and the x-axis of the parking coordinate system.

[0158] Based on the above, the five-segment distance It satisfies the following formula 18.

[0159] (Equation 18).

[0160] in, express and The difference between them express and The difference between them express and The difference between them.

[0161] Based on the above, a system of nonlinear equations is constructed using equations 13-18, and a five-segment distance is used. The minimum constraint is used to optimize the solution, and the planned path from the second target node to the termination point is obtained.

[0162] Based on the above technical solutions, by using RS curve planners or geometric path planners, combined with node expansion and path splicing strategies, parking paths that balance smoothness, obstacle avoidance capabilities, and high efficiency can be generated, thereby improving the quality of parking paths.

[0163] In one optional implementation, the above-mentioned determination of the vehicle's parking path based on the optimal path may specifically include: the vehicle generating a STC region based on a quadtree map and determining the ideal path point corresponding to each path point on the optimal path within the STC region. Then, the vehicle smooths the optimal path with the objective of minimizing the sum of the differences between each path point on the smoothed optimal path and its corresponding ideal path point, and minimizing the smoothing loss of adjacent path points on the smoothed optimal path. Finally, the vehicle can determine the smoothed optimal path as its parking path.

[0164] The STC area is a T-shaped area bounded by the target parking space and the obstacle-free area defined by the quadtree map. The ideal waypoint is close to the center line of the STC area.

[0165] In one example, we assume the optimal path .in, Represents the optimal path The first in Path points, It is a natural number. For the optimal path... Path points in Its ideal path point in the STC region is , This represents a subregion of the STC region.

[0166] Based on the above, the vehicle can use an optimization solver to determine the target smooth optimal path based on each path point on the optimal path and the ideal path point corresponding to each path point in the STC region. Among them, the minimum length of the target smooth optimal path. It satisfies the following formula 19.

[0167] (Equation 19).

[0168] in, express Smoothed points. This represents the sum of the differences between each path point on the smoothed optimal path and its corresponding ideal path point. This represents the smoothing loss between adjacent path points on the optimal path after smoothing. This indicates the preset weighting coefficient.

[0169] Based on the above technical solution, the center line of the STC area is used as a reference for the ideal path to ensure that the path points are close to the center of the channel, avoid getting too close to obstacles, reduce the risk of collision, and improve path safety.

[0170] In one optional implementation, the method of determining the vehicle's parking path based on the smoothed optimal path can be specifically implemented as follows: the vehicle can find target path point pairs on the smoothed optimal path whose distance is less than a preset distance threshold and whose directional angle difference is less than a preset angle. Then, if the turning angle between the target path point pairs is less than a preset turning angle, the path between the target path point pairs is trimmed. Finally, the trimmed optimal path can be determined as the vehicle's parking path.

[0171] In this embodiment, the preset distance threshold and preset angle are determined based on the Euclidean distance between the starting point and the ending point. The Euclidean distance between the starting point and the ending point... It is determined that the following formula 20 is satisfied.

[0172] (Equation 20).

[0173] in, Indicates the starting point position. Indicates the location of the termination point.

[0174] In one example, the vehicle can determine whether the number of gear shifts in the smoothed optimal path exceeds a shift threshold. If it does not, no path pruning is performed, i.e., the smoothed optimal path is preserved. If it does exceed the threshold, a target path pair with a distance less than a preset distance threshold and a directional angle difference less than a preset angle is found. Then, if the steering angle between the target path pair is less than a preset steering angle, the path between the target path pair is pruned; if the steering angle between the target path pair is greater than the preset steering angle, no path pruning is performed.

[0175] Based on the above technical solution, the target path point pairs can be filtered through triple constraints of distance threshold, direction angle difference, and steering angle on the smoothed optimal path, and redundant paths between two points can be trimmed to generate a simpler parking path. This method improves parking efficiency and smoothness by removing invalid curves or detours in the path.

[0176] In one optional implementation, the method provided in this application may further include: constructing a raster map, a quadtree map, a Venn diagram, and a distance map. The following detailed description uses 1. constructing a raster map, 2. constructing a distance map, 3. constructing a Venn diagram, and 4. constructing a quadtree map as examples.

[0177] 1. Construct a raster map.

[0178] In some embodiments, the vehicle can convert the coordinates of surrounding obstacles in the odometry coordinate frame (ODOM) to coordinates in the parking coordinate system. Furthermore, the vehicle can create an initial grid map, i.e., an empty grid map. Then, the vehicle can map obstacle information from the parking coordinate system onto the initial grid map to obtain a grid map containing obstacle information, wherein the grids containing obstacles are occupied grids.

[0179] 2. Construct a distance map.

[0180] In some embodiments, the vehicle can establish an initial distance map using the method described above. Then, combining the grid map described above, for each grid cell in the grid map, the vehicle can calculate the distance from that grid cell to each occupied grid cell, and store the minimum distance in the initial distance map to obtain the final distance map.

[0181] For example, if the distance from grid A to the occupied grid B is 1, and the distance from grid A to the occupied grid C is 2, then 1 is stored in the initial distance map.

[0182] 3. Construct the Veno diagram.

[0183] In some embodiments, the vehicle can extract the outer contour boundary of each obstacle based on obstacle information in the parking coordinate system, then use the midpoint of each obstacle as a seed point, and calculate the perpendicular bisector between each pair of seed points, i.e. the edge of the Veno graph, to obtain the Veno graph.

[0184] 4. Construct a quadtree map.

[0185] In some embodiments, the vehicle can establish an initial quadtree map using the method described above. Then, the vehicle can map obstacle information from the parking coordinate system onto the initial quadtree map to obtain a quadtree map containing obstacle information (hereinafter referred to as the first quadtree map). Subsequently, based on the obstacle information in the first quadtree map and the minimum area resolution, the vehicle can recursively divide the area containing obstacles into four sub-regions until the minimum area resolution is reached or no obstacles exist.

[0186] Specifically, the process first checks whether each region contains obstacles. For regions containing obstacles, if the current region size is larger than the minimum region resolution, the region is divided into four sub-regions. The steps of checking for obstacles and dividing the region are repeated for each sub-region until the region containing obstacles reaches the minimum region resolution or does not contain obstacles. For regions that do not contain obstacles, the division process stops as there are no obstacles.

[0187] Based on the above technical means, this application can construct grid maps, quadtree maps, Veno maps and distance maps based on obstacle information, thereby providing support for path planning maps and calculating the path cost of each node.

[0188] Figure 4 This is a schematic diagram of the structure of a parking path planning device provided in an embodiment of this application, as shown below. Figure 4 As shown, the parking path planning device includes: a first determining unit 401, a second determining unit 402, and a planning unit 403.

[0189] The first determining unit 401 is used to determine the target location point based on the obstacle information around the target parking space; wherein, the target location point is the projection point of the center point of the rear axle of the vehicle after parking onto the target parking space.

[0190] The second determining unit 402 is used to determine the target planning map based on the environmental complexity of parking. When the environmental complexity is greater than or equal to a preset complexity threshold, the target planning map is a grid map; when the environmental complexity is less than the preset complexity threshold, the target planning map is a quadtree map. Both the grid map and the quadtree map are constructed based on obstacle information.

[0191] Planning unit 403 is used to plan the parking path of the vehicle based on the target location point and the target planning map.

[0192] In one possible approach, the first determining unit 401 is specifically used to: determine an ideal position point, wherein the ideal position point is, under ideal conditions, the projection point of the center point of the rear axle of the vehicle onto the target parking space; determine the available parking space for the vehicle based on the parking space boundary and obstacle information of the target parking space; and determine the target position point based on the ideal position point and the available parking space.

[0193] In one possible embodiment, the first determining unit 401 further includes a calculation subunit and a processing subunit. The calculation subunit calculates the offset distances between the four sides of the rectangle and their corresponding parking space sides, provided that the length and width of the rectangle are greater than the length and width of the vehicle, respectively. The processing subunit adjusts the ideal position point based on the offset distances to obtain the target position point.

[0194] In one possible approach, the first determining unit 401 is specifically used to determine the target location point as the ideal location point when the length of the rectangle is less than the length of the vehicle, or the width of the rectangle is less than the width of the vehicle.

[0195] In one possible implementation, planning unit 403 further includes a planning subunit, a selection subunit, and a determination subunit. The planning subunit is used to plan multiple paths on the target planning map using the Hybrid A* algorithm, starting from the vehicle's current rear axle center point and ending at the target location point. The selection subunit is used to select the optimal path from the multiple paths based on the path lengths, the number of reversing segments, and the number of curvature changes. The determination subunit is used to determine the vehicle's parking path based on the optimal path.

[0196] In one possible approach, the planning sub-unit is specifically used to generate multiple feasible motion trajectories starting from the first target node, based on the vehicle's maximum permissible steering angle, during each node expansion process. The intersection of these multiple motion trajectories with the first grid line of the target planning map is determined as a first candidate node. The first candidate node that meets preset conditions is determined as a second candidate node, where the preset conditions constrain the vehicle to have no collision risk with obstacles during its passage through the first candidate node; the collision risk is determined based on a distance map. The path cost of the second candidate node is calculated based on a Venn diagram, where the Venn diagram and distance map are determined based on obstacle information. The second candidate node with the minimum path cost is determined as the second target node. Based on the second target node, a path from the starting point to the ending point is generated.

[0197] In the initial node expansion process, the first target node is the starting point; if the second target node is the ending point, the node expansion ends; if the second target node is not the ending point, the second target node becomes the first target node in the next node expansion process.

[0198] In one possible approach, the sub-unit is specifically used to generate a STC region based on a quadtree map, where the STC region is a T-shaped area enclosed by the target parking space and the obstacle-free area defined by the quadtree map. The ideal path point corresponding to each path point on the optimal path is determined within the STC region; the ideal path point is close to the centerline of the STC region. The goal of smoothing the optimal path is to minimize the sum of the differences between each path point on the smoothed optimal path and its corresponding ideal path point, and to minimize the smoothing loss of adjacent path points on the smoothed optimal path. Based on the smoothed optimal path, the parking path for the vehicle is determined.

[0199] Figure 5 This is a block diagram of an electronic device provided in an embodiment of this application. (For example...) Figure 5 As shown, the electronic device includes, but is not limited to, a processor 501 and a memory 502.

[0200] The memory 502 described above is used to store the executable instructions of the processor 501. It is understood that the processor 501 is configured to execute instructions to implement the vehicle control method in the above embodiments.

[0201] It should be noted that those skilled in the art will understand that Figure 5 The electronic device structure shown does not constitute a limitation on the electronic device; the electronic device may include, but is not limited to, other electronic devices. Figure 5 This may indicate more or fewer components, or combinations of certain components, or different component arrangements.

[0202] Processor 501 is the control center of the electronic device. It connects various parts of the electronic device via various interfaces and lines. By running or executing software programs and / or modules stored in memory 502, and by calling data stored in memory 502, it performs various functions and processes data, thereby providing overall monitoring of the electronic device. Processor 501 may include one or more processing units. Optionally, processor 501 may integrate an application processor and a modem processor. The application processor mainly handles the operating system, user interface, and applications, while the modem processor mainly handles wireless communication. It is understood that the modem processor may not be integrated into processor 501.

[0203] The memory 502 can be used to store software programs and various data. The memory 502 may primarily include a program storage area and a data storage area. The program storage area may store the operating system, application programs required by at least one functional module (such as a determination unit, processing unit, etc.), etc. Furthermore, the memory 502 may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0204] In an exemplary embodiment, a computer-readable storage medium including instructions is also provided, such as a memory 502 including instructions, which can be executed by a processor 501 of an electronic device to implement the methods in the above embodiments.

[0205] In actual implementation, Figure 4 The functions of the first determining unit 401, the second determining unit 402, and the planning unit 403 can all be derived from... Figure 5 The processor 501 calls the computer program stored in the memory 502 to implement the process. The specific execution process can be found in the description of the method section in the previous embodiment, and will not be repeated here.

[0206] Optionally, the computer-readable storage medium may be a non-transitory computer-readable storage medium, such as a read-only memory (ROM), random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device.

[0207] In an exemplary embodiment, this application also provides a computer program product including one or more instructions, which can be executed by a processor 501 of an electronic device to perform the methods described above.

[0208] It should be noted that when one or more instructions in the computer-readable storage medium or computer program product are executed by the processor of an electronic device, they implement the various processes of the above method embodiments and achieve the same technical effect as the above method. To avoid repetition, they will not be described again here.

[0209] Through the above description of the embodiments, those skilled in the art can clearly understand that, for the sake of convenience and brevity, only the division of the above functional modules is used as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.

[0210] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another apparatus, or some features may be ignored or not executed. Furthermore, the mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0211] The units described as separate components may or may not be physically separate. A component shown as a unit can be one or more physical units; that is, it can be located in one place or distributed in multiple different locations. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0212] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0213] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a readable storage medium. Based on this understanding, the technical solutions of the embodiments of this application, essentially, or the parts that contribute to the prior art, or all or part of the technical solutions, can be embodied in the form of a software product. This software product is stored in a storage medium and includes several instructions to cause a device (which may be a microcontroller, chip, etc.) or processor to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, ROM, RAM, magnetic disks, or optical disks.

[0214] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A parking path planning method, characterized in that, The method includes: Based on obstacle information around the target parking space, a target location point is determined, wherein the target location point is the projection point of the center point of the rear axle of the vehicle after parking in the target parking space. When the environmental complexity of the target parking space is high, the target planning map is determined to be a raster map; When the environmental complexity of the target parking space is low, the target planning map is determined to be a quadtree map; both the grid map and the quadtree map are constructed based on the obstacle information. On the target planning map, the Hybrid A* algorithm is used to plan multiple paths starting from the current rear axle center point of the vehicle and ending at the target location point. Based on the length of the multiple paths, the number of reversing segments, and the number of curvature changes, the optimal path is selected from the multiple paths. Based on the optimal path, the parking path of the vehicle is determined.

2. The method according to claim 1, characterized in that, The process of determining the target location point based on obstacle information around the target parking space includes: Determine the ideal location point, wherein the ideal location point is the projection point of the center point of the rear axle of the vehicle in the target parking space under ideal conditions; Based on the parking space boundary of the target parking space and the obstacle information, the available parking space for the vehicle is determined; The target location is determined based on the ideal location and the available parking space.

3. The method according to claim 2, characterized in that, The spatial boundary of the parking space is a rectangle; the four sides of the rectangle correspond one-to-one with the four sides of the parking space boundary; determining the target location point based on the ideal location point and the parking space includes: If the length and width of the rectangle are greater than the length and width of the vehicle, calculate the offset distance between the four sides of the rectangle and the corresponding parking space side; The target position point is obtained by adjusting the ideal position point based on the offset distance.

4. The method according to claim 3, characterized in that, The process of determining the target location point based on the ideal location point and the available parking space further includes: If the length of the rectangle is less than the length of the vehicle, or the width of the rectangle is less than the width of the vehicle, the target location point is determined as the ideal location point.

5. The method according to claim 1, characterized in that, The process of planning multiple paths on the target planning map using the HybridA* algorithm, starting from the current rear axle center point of the vehicle and ending at the target location point, includes: During each node expansion process, based on the vehicle's maximum permissible steering angle, multiple feasible motion trajectories starting from the first target node are generated; The intersection points of the multiple motion trajectories with the first grid line of the target planning map are determined as the first candidate nodes; The first candidate node that meets the preset conditions is determined as the second candidate node, wherein the preset conditions are used to constrain the vehicle to have no risk of collision with obstacles during the process of passing through the first candidate node; the collision risk is determined based on the distance map. Based on the Veno map, the path cost of the second candidate node is calculated; wherein the Veno map and the distance map are determined based on the obstacle information; The second candidate node with the lowest path cost is determined as the second target node; Based on the second target node, generate a path from the starting point to the ending point; In the initial node expansion process, the first target node is the starting point; if the second target node is the ending point, the node expansion ends; if the second target node is not the ending point, the second target node becomes the first target node in the next node expansion process.

6. The method according to claim 5, characterized in that, The calculation of the path cost of the second candidate node based on the Venn diagram includes: Based on the Venn diagram, estimate the path cost between the second candidate node and the termination point; Based on the changes in travel direction, steering wheel angle, and gear shifting between the first target node and the second candidate node, the path cost between the first target node and the second candidate node is determined. The sum of the cumulative path cost from the starting point to the first target node, the path cost between the first target node and the second candidate node, and the path cost between the second candidate node and the ending point is determined as the path cost of the second candidate node.

7. The method according to claim 6, characterized in that, The step of estimating the path cost between the second candidate node and the termination point based on the Venn diagram includes: Determine the maximum and minimum distances between the second candidate node and the Vino edge in the Vino graph; Based on the maximum distance and the minimum distance, a first cost is calculated, wherein a smaller first cost indicates that the second candidate node is closer to the Vino edge; The second cost is calculated based on the distance between the second candidate node and the obstacle; The third cost is determined based on the distance between the second candidate node and the termination point; The fourth cost is determined based on the distance between the second candidate node and the starting point; The fifth cost is determined based on the distance between the second candidate node and the guide point of the target parking space; Based on the first cost, the second cost, the third cost, the fourth cost, and the fifth cost, the path cost between the second candidate node and the termination point is estimated.

8. The method according to claim 5, characterized in that, The step of generating the path from the starting point to the ending point based on the second target node includes: Using an RS curve planner and / or a geometric path planner, a planned path from the second target node to the termination point is generated; The first path is obtained by concatenating the expansion paths between the first and second target nodes in the previous node expansion process. The first path is concatenated with the planned path to obtain the path from the starting point to the ending point.

9. The method according to claim 8, characterized in that, The geometric path planner uses a two-stage path planning algorithm to generate a planned path from the second target node to the termination point; The two-stage path planning algorithm aims to minimize the sum of the first arc from the second target node to the first intermediate node and the second arc from the first intermediate node to the termination point, and optimizes the solution by ensuring that the tangent vectors of the circles containing the first arc and the circles containing the second arc are continuous at the first intermediate node.

10. The method according to claim 8, characterized in that, The geometric path planner uses a five-segment path planning algorithm to generate a planned path from the second target node to the termination point; The five-segment path planning algorithm optimizes the solution by minimizing the straight line segment between the second target node and the second intermediate node, the arc segment between the second intermediate node and the third intermediate node, the arc segment between the third intermediate node and the fourth intermediate node, the arc segment between the fourth intermediate node and the fifth intermediate node, and the distance between the fifth intermediate node and the termination point. The position of the third intermediate node and the radii of the circles containing the arc segments from the second intermediate node to the third intermediate node, the arc segments from the third intermediate node to the fourth intermediate node, and the arc segments from the fourth intermediate node to the fifth intermediate node are predetermined.

11. The method according to claim 1, characterized in that, Determining the parking path of the vehicle based on the optimal path includes: Based on the quadtree map, an STC region is generated, wherein the STC region is a T-shaped region enclosed by the target parking space and the obstacle-free area defined by the quadtree map; Determine the ideal path point corresponding to each path point on the optimal path in the STC region; the ideal path point is close to the centerline of the STC region; The goal is to smooth the optimal path by minimizing the sum of the differences between each path point on the smoothed optimal path and its corresponding ideal path point, and by minimizing the smoothing loss of adjacent path points on the smoothed optimal path. The parking path of the vehicle is determined based on the smoothed optimal path.

12. The method according to claim 11, characterized in that, Determining the parking path of the vehicle based on the smoothed optimal path includes: Find target path point pairs on the smoothed optimal path whose distance is less than a preset distance threshold and whose direction angle difference is less than a preset angle. If the turning angle between the target path point pairs is less than a preset turning angle, the path between the target path point pairs will be cut off. The optimal path after trimming is determined as the parking path for the vehicle.

13. A parking path planning device, characterized in that, The device includes: The first determining unit is used to determine the target location point based on obstacle information around the target parking space; wherein, the target location point is the projection point of the center point of the rear axle of the vehicle after parking in the target parking space; The second determining unit is used to determine the target planning map as a grid map when the environmental complexity of the target parking space is high, and to determine the target planning map as a quadtree map when the environmental complexity of the target parking space is low; both the grid map and the quadtree map are constructed based on the obstacle information. The planning unit is used to plan multiple paths on the target planning map using the Hybrid A* algorithm, starting from the current rear axle center point of the vehicle and ending at the target location point. The processing unit is configured to select the optimal path from the multiple paths based on the length of the multiple paths, the number of reversing segments, and the number of curvature changes; The third determining unit is used to determine the parking path of the vehicle based on the optimal path.

14. A vehicle, characterized in that, The vehicle is equipped with the device as described in claim 13.

15. An electronic device, characterized in that, include: processor; Memory used to store the processor's executable instructions; The processor is configured to execute the instructions to implement the method as described in any one of claims 1-12.

Images