A parking path planning method and device, storage medium and electronic equipment

Abstract

The application relates to a parking path planning method and device, a storage medium and an electronic device, which comprises the following steps: acquiring surrounding environment information from a target parking space to a current position of a vehicle, and obtaining obstacle information, a starting point and an ending point according to the surrounding environment information; based on a preset vehicle kinematics model, performing first path searching according to the obstacle information, the starting point and the ending point to obtain an initial extended tree; on the basis of the initial extended tree, performing second path searching according to the obstacle information, the starting point and the ending point and a preset RRT algorithm to obtain a random extended tree; wherein the random extended tree comprises a plurality of nodes, node information of each node of the random extended tree comprises position coordinates and a heading angle corresponding to the node; and a parking planning path is obtained according to the random extended tree. The application can solve the problem that the search of a parking trajectory is prone to failure in the current scene with narrow parking spaces and many obstacles.

Description

Technical Field

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

[0002] Automated parking is a vehicle assistance technology that uses sensors and computer algorithms to identify target parking spaces and control the vehicle to perform autonomous driving operations, helping the driver park the vehicle in the desired space. The automated parking system measures the size of the target parking space and calculates the optimal parking path and steering angle based on the vehicle's dimensions and environmental conditions. It then automatically controls the steering wheel, accelerator, and brakes to achieve automatic parking. Automated parking technology reduces driver stress and skill requirements when parking, improves parking efficiency and safety, and reduces the risk of collisions with surrounding objects. Automated parking technology is currently widely used by many car brands, and with its continuous development and popularization, it will become an important part of future driving.

[0003] The Rapidly-exploring Random Tree (RRT) algorithm for automated parking is a path planning algorithm based on a tree-structured data structure. It can search for effective paths in a high-dimensional space and find the optimal path from the starting point to the destination. RRT is an incremental sampling search algorithm, which has the advantage of tending to expand to open, unexplored regions and possessing probabilistic completeness; given enough time and numerous iterations, all regions will be explored. However, RRT is a purely random search algorithm and is insensitive to environmental type. When the space contains many obstacles or narrow passage constraints, the algorithm's convergence speed is slow, and its efficiency drops significantly. Therefore, in scenarios with narrow parking spaces and many obstacles, parking trajectory searches based on the RRT algorithm are prone to failure. Summary of the Invention

[0004] The purpose of this application is to propose a parking path planning method and device to solve the problem that parking trajectory search is prone to failure in scenarios with narrow parking spaces and many obstacles.

[0005] To achieve the above objectives, embodiments of this application provide a parking route planning method, the method comprising:

[0006] Obtain surrounding environmental information from the target parking space to the vehicle's current position, and obtain obstacle information, start point, and end point based on the surrounding environmental information; wherein, the start point is the location of the target parking space, and the end point is the vehicle's current position;

[0007] Based on a preset vehicle kinematics model, an initial extended tree is obtained by performing a first path search according to the obstacle information, the starting point, and the ending point; wherein, the initial extended tree includes multiple nodes, and the node information of each node includes the position coordinates and heading angle corresponding to the node;

[0008] Based on the initial extended tree, a second path search is performed according to the obstacle information, the start and end points, and the preset RRT algorithm to obtain a random extended tree; wherein, the random extended tree includes multiple nodes, and the node information of each node includes the position coordinates and heading angle corresponding to the node;

[0009] The parking planning path is obtained based on the random expansion tree.

[0010] The method provided in this application, in order to adapt to the uneven distribution of obstacle density in wide driving lanes and narrow parking spaces, uses the target parking space as the starting point and the current vehicle position as the ending point to perform a reverse path search. Furthermore, during the search process, a first path search is performed based on the vehicle's kinematics model, pre-generating an initial expansion tree. A second path search is then performed based on this initial expansion tree combined with the RRT algorithm. This accelerates the growth process of the random expansion tree, avoiding the problem of the RRT algorithm struggling to grow within the parking space during reverse search, which can even lead to planning failure. Pre-generating the initial expansion tree reduces the number of invalid expansions, preventing frequent collision detection and invalid expansions caused by random expansion tree growth in narrow parking spaces, thus greatly improving the search speed. This effectively solves the problem of parking trajectory search easily failing in scenarios with narrow parking spaces and many obstacles.

[0011] In some implementations, the initial path search based on a preset vehicle kinematics model, according to the starting point and the ending point, specifically includes:

[0012] Using the starting point as the root node of the initial expansion tree, nodes are expanded in multiple expansion directions, with each expansion direction corresponding to a preset wheel angle.

[0013] When expanding a node in any direction, the vehicle moves a preset step from the current node according to a preset wheel angle and performs collision prediction based on the obstacle information. If no collision occurs after the movement, the new position is taken as a new node in the expansion direction, and it is determined whether to end the node expansion in that direction. If a collision occurs after the movement, the vehicle moves a preset step in a straight line from the current node, and the new position is taken as a new node in the expansion direction, and it is determined whether to end the node expansion in that direction.

[0014] If the position of the new node meets the preset conditions, the expansion ends; if the position of the new node does not meet the preset conditions, the expansion continues in the same direction with the new node as the current node.

[0015] In some implementations, the first path search based on the starting point and the ending point further includes:

[0016] When a new node is obtained, the first position coordinates and first heading angle of the parent node of the new node, the first wheel angle and moving speed of the vehicle from the parent node to the new node are obtained, and the position coordinates and heading angle of the new node are calculated based on the preset vehicle kinematics model according to the first position coordinates, first heading angle, first wheel angle and moving speed.

[0017] In some implementations, the preset conditions include:

[0018] When a vehicle moves to a new node, it meets the condition of leaving the target parking space; wherein, the condition for leaving the target parking space is that the vehicle body completely leaves the parking space line of the target parking space, or the front of the vehicle touches the driving lane boundary above the parking space.

[0019] In some implementations, the preset wheel angles corresponding to the plurality of extension directions are as follows:

[0020] φ max , -φ max ;

[0021] Where, φ mac This is the preset maximum wheel angle of the vehicle.

[0022] In some implementations, the step of performing a second path search to obtain a randomized expanded tree based on the initial expanded tree, according to the obstacle information, the start and end points, and a preset RRT algorithm, specifically includes:

[0023] Multiple nodes of the initial expanded tree are used as multiple base nodes of the random expanded tree, and the random expanded tree is grown at least once based on the multiple base nodes until there is a node in the random expanded tree whose distance cost to the destination is less than a preset threshold. Then the growth stops and the second path search ends.

[0024] In some implementations, the distance cost between any node in the randomly expanded tree and the endpoint is determined as follows:

[0025] If the vehicle moves to any node of the random expansion tree and does not enter the driving lane, then obtain the second heading angle of the vehicle when it is located at that node, the first distance between the vehicle and the center line of the driving lane, and the tilt angle of the center line of the driving lane, and determine the distance cost between that node and the destination based on the second heading angle, the tilt angle of the center line of the driving lane, and the first distance.

[0026] If the vehicle has entered the driving lane when it moves to any node of the randomly expanded tree, then obtain the third heading angle and second position coordinates of the vehicle when it is located at that node, the destination heading angle and destination position coordinates of the vehicle when it is located at the destination, and determine the distance cost between the arbitrary node and the destination based on the third heading angle, second position coordinates, destination heading angle, and destination position coordinates.

[0027] In some implementations, each growth of the randomly expanding tree includes:

[0028] Random sampling is performed to obtain temporary sampling points, and a node is selected from the existing nodes of the random expansion tree as the nearest neighbor node of the temporary sampling point.

[0029] From the nearest neighbor node, move a preset step length in multiple directions according to multiple preset wheel turning angles, and perform collision prediction based on the obstacle information;

[0030] If no collision occurs after moving in any direction, the position after moving in any direction will be used as a candidate child node.

[0031] If there is at least one candidate child node, then select the candidate child node with the smallest distance cost from the endpoint from the at least one candidate child node, and use it as a new node in the random expansion tree, with its parent node being the nearest neighbor node, thus completing one growth cycle.

[0032] In some implementations, selecting a candidate child node from the at least one candidate child node that has the minimum distance cost to the endpoint specifically includes:

[0033] If the vehicle is located at any candidate sub-node and has not entered the driving lane, then obtain the fourth heading angle of the vehicle when it is located at that candidate sub-node, the second distance between the vehicle and the center line of the driving lane, and the tilt angle of the center line of the driving lane. Then, determine the distance cost between the candidate sub-node and the destination based on the fourth heading angle, the second distance, and the destination heading angle.

[0034] If a vehicle has entered the driving lane when it is located at any candidate sub-node, then obtain the fifth heading angle and third position coordinates when the vehicle is located at that candidate sub-node, and the destination heading angle and destination position coordinates when the vehicle is located at the destination. Then, determine the distance cost between the candidate sub-node and the destination based on the fifth heading angle, third position coordinates, destination heading angle, and destination position coordinates.

[0035] In some implementations, the plurality of preset wheel angles include:

[0036]

[0037] Where, φ max This is the preset maximum wheel angle of the vehicle.

[0038] This application also provides a parking path planning device, including a module for performing the parking path planning method as described above.

[0039] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the parking path planning method described above.

[0040] This application also provides an electronic device, including a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the parking path planning method described above. Attached Figure Description

[0041] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the accompanying drawings required in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0042] Figure 1 This is a flowchart of a parking path planning method in one embodiment of this application.

[0043] Figure 2 This is a schematic diagram of the parking space coordinate system in one embodiment of this application.

[0044] Figure 3 This is a schematic diagram of the path a vehicle takes to exit a target parking space, as shown in one embodiment of this application.

[0045] Figure 4 This is a flowchart of the first path search in one embodiment of this application.

[0046] Figure 5 This is a schematic diagram of a parking path planning device in one embodiment of this application. Detailed Implementation

[0047] The detailed description of the accompanying drawings is intended to illustrate the present preferred embodiments of this application and is not intended to represent only the forms in which this application can be implemented. It should be understood that the same or equivalent functions can be achieved by different embodiments intended to be included within the spirit and scope of this application.

[0048] One embodiment of this application provides a parking path planning method, see reference. Figure 1 The method in this embodiment includes the following steps:

[0049] Step S100: Obtain the surrounding environment information from the target parking space to the current position of the vehicle, and obtain obstacle information, start point and end point based on the surrounding environment information; wherein, the start point is the position of the target parking space and the end point is the current position of the vehicle.

[0050] Specifically, this embodiment collects environmental data of the driving lane and parking space using an onboard camera, and performs real-time positioning using an onboard RTK device to obtain the UTM coordinates x of the vehicle and surrounding obstacles. u y u Perform a coordinate rotation transformation to obtain the x-coordinates in the parking space coordinate system. p y p The parking space coordinate system is a right-handed coordinate system with the origin P as the origin and the Y-axis pointing vertically outward from the target parking space, such as... Figure 2 As shown in formula (1), x pu y pu Let θ be the UTM coordinates of point P. pu The azimuth angle of the target parking space under the UTM;

[0051]

[0052] In this embodiment, when the geometric center of the target parking space coincides with the geometric center of the vehicle, the location of the rear axle center point of the vehicle is defined as the starting point P, and the ending point is the current location of the rear axle center point of the vehicle.

[0053] Step S200: Based on the preset vehicle kinematics model, perform the first path search to obtain an initial extended tree according to the obstacle information, starting point and ending point; wherein, the initial extended tree includes multiple nodes, and the node information of each node includes the position coordinates and heading angle corresponding to the node.

[0054] Specifically, the vehicle kinematics model in this embodiment can be a bicycle model, from which the following formula (2) can be obtained;

[0055]

[0056] in, These represent the rate of change of the x-coordinate of the rear axle center, the rate of change of the y-coordinate of the rear axle center, and the rate of change of the heading angle, respectively; v is the velocity of the rear axle center; θ is the velocity of the rear axle center. c φ is the heading angle before the vehicle moves, and φ is the front wheel steering angle during the movement.

[0057] It should be noted that the first path search in this embodiment is based on the parking space coordinate system. During the path search process, more nodes are searched starting from the starting point as the root node. An edge is formed between the nodes. Finally, connecting multiple edges can yield a path to drive out of the target parking space, which is the initial extended tree. The movement law of the vehicle is related to the vehicle kinematics model. Therefore, in this embodiment, the vehicle movement is predicted based on the vehicle kinematics model during the path search process. For path planning, it is necessary to obtain the position coordinates and heading angle of each trajectory point on the path to guide the autonomous driving of the vehicle. Specifically, in this embodiment, the position coordinates and heading angle of each node can be obtained based on formula (2).

[0058] Step S300: Based on the initial extended tree, a second path search is performed according to the obstacle information, the starting point and the ending point, and the preset RRT algorithm to obtain a random extended tree; wherein, the random extended tree includes multiple nodes, and the node information of each node includes the position coordinates and heading angle corresponding to the node.

[0059] Specifically, in this embodiment, the root node of the random expansion tree is the starting point. The second path search in this embodiment is still based on the parking space coordinate system. The Rapidly-exploring Random Tree (RRT) algorithm is a commonly used algorithm for robot path planning. It searches for the optimal path by continuously expanding the tree structure. When performing path search, the conventional RRT algorithm needs to initialize the random expansion tree, randomly select a starting point as the root node of the random expansion tree, and add it to the tree. The tree structure is then expanded based on the root node. However, in this embodiment, the initial expansion tree is used as the basis. That is to say, before expanding the random expansion tree, the random expansion tree already has multiple nodes, which speeds up the growth process of the random expansion tree and avoids frequent collision detection and invalid expansion caused by the growth of the random expansion tree in narrow parking spaces, thus greatly improving the search speed.

[0060] Step S400: Obtain the parking planning path based on the random expansion tree.

[0061] Specifically, after obtaining the random expansion tree, starting from the leaf node of the random expansion tree (the distance cost between the leaf node and the destination is less than a preset threshold), the path from the starting point to the destination is obtained by tracing back along the generated edges (two nodes connected to form an edge). This path is the parking planning path. The node information of each node includes the position coordinates and heading angle of the corresponding node. Based on the position coordinates and heading angles of multiple nodes on the parking planning path, the autonomous driving of the vehicle can be controlled.

[0062] The method provided in this application, in order to adapt to the uneven distribution of obstacle density in wide driving lanes and narrow parking spaces, uses the target parking space as the starting point and the current vehicle position as the ending point to perform a reverse path search. Furthermore, during the search process, a first path search is performed based on the vehicle's kinematics model, pre-generating an initial expansion tree. A second path search is then performed based on this initial expansion tree combined with the RRT algorithm. This accelerates the growth process of the random expansion tree, avoiding the problem of the RRT algorithm struggling to grow within the parking space during reverse search, which can even lead to planning failure. Pre-generating the initial expansion tree reduces the number of invalid expansions, preventing frequent collision detection and invalid expansions caused by random expansion tree growth in narrow parking spaces, thus greatly improving the search speed. This effectively solves the problem of parking trajectory search easily failing in scenarios with narrow parking spaces and many obstacles.

[0063] In some embodiments, step S200 specifically includes:

[0064] Using the starting point as the root node of the initial expansion tree, nodes are expanded in multiple expansion directions, with each expansion direction corresponding to a preset wheel angle.

[0065] When expanding a node in any direction, the vehicle moves a preset step from the current node according to a preset wheel angle and performs collision prediction based on the obstacle information. If no collision occurs after the movement, the new position is taken as a new node in the expansion direction, and it is determined whether to end the node expansion in that direction. If a collision occurs after the movement, the vehicle moves a preset step in a straight line from the current node, and the new position is taken as a new node in the expansion direction, and it is determined whether to end the node expansion in that direction.

[0066] If the position of the new node meets the preset conditions, the expansion ends; if the position of the new node does not meet the preset conditions, the expansion continues in the same direction with the new node as the current node.

[0067] Specifically, the movement step refers to the distance the rear axle center of the vehicle moves by one step. The length of the edge formed by connecting the parent node and the child node is one movement step. In this embodiment, the rear axle center of the vehicle is taken as the location point of the vehicle.

[0068] The collision prediction refers to predicting whether a vehicle will collide with an obstacle after moving in a certain direction, based on vehicle body parameters, vehicle speed, and obstacle information; the obstacle information includes the obstacle's structure, location, speed, and direction of movement.

[0069] Expanding the leaf nodes can yield multiple paths that allow vehicles to exit the parking space. Each path starts from the root node. For example... Figure 3 As shown, Figure 3 The nodes were expanded in six directions, resulting in six paths that allow vehicles to exit the parking space. The search process for each path can be found in [reference needed]. Figure 4 During the search of a path in each expansion direction, the number of straight-line steps (straight_count) and the number of turning steps (turn_count) are recorded. After obtaining a new node, it is determined whether the vehicle has left the target parking space. If not, the expansion continues to add new nodes; if so, the process ends, and the path generated according to the number of straight-line steps (straight_count) and the number of turning steps (turn_count) is added to the initial expansion tree.

[0070] In some embodiments, the first path search based on the start point and the end point further includes:

[0071] When a new node is obtained, the first position coordinates and first heading angle of the parent node of the new node, the first wheel angle and moving speed of the vehicle from the parent node to the new node are obtained, and the position coordinates and heading angle of the new node are calculated based on the preset vehicle kinematics model according to the first position coordinates, first heading angle, first wheel angle and moving speed.

[0072] Specifically, the position coordinates and heading angle of each new node can be based on the vehicle kinematics model, i.e., the content of the aforementioned formula (2), according to the moving speed v of the vehicle's rear axle center and the heading angle θ of the current node. c The rate of change of the horizontal axis can be calculated from the first wheel rotation angle φ during the movement. Rate of change of the ordinate Rate of change of heading angle Based on the rate of change and the initial position coordinates before the movement (including the x-coordinate) c y-axis c First heading angle θ c This allows us to calculate the x-coordinate, y-coordinate, and heading angle of the new node after it has been moved. The position coordinates include both the x-coordinate and y-coordinate.

[0073] In some embodiments, the preset conditions include:

[0074] When a vehicle moves to a new node, it meets the condition of having exited the target parking space; wherein, the determination condition for exiting the target parking space is that the vehicle body completely exits the parking space line of the target parking space, or the front of the vehicle touches the lane boundary above the parking space, for example. Figure 3 As shown.

[0075] In some embodiments, to balance search efficiency and search success rate, the plurality of expansion directions include six directions, and the plurality of preset wheel angles corresponding to these six directions may be, but are not limited to, the following:

[0076] φ max , -φ max ;

[0077] Where, φ max The preset maximum wheel turning angle of the vehicle is defined as follows: if the left is the positive direction, when the steering wheel is turned to the left, the wheel turns to the left, and the turning angle is positive. Conversely, when the steering wheel is turned to the right, the wheel turns to the right, and the turning angle is negative.

[0078] In some embodiments, step S300 specifically includes:

[0079] Multiple nodes of the initial expanded tree are used as multiple base nodes of the random expanded tree, and the random expanded tree is grown at least once based on the multiple base nodes until there is a node in the random expanded tree whose distance cost to the destination is less than a preset threshold. Then the growth stops and the second path search ends.

[0080] Specifically, the steps involve growing a random expansion tree based on the initial expansion tree. By continuously generating nodes randomly and attempting to connect nodes to the random expansion tree, a path that can connect the start and end points is sought. During the process of connecting nodes, it is necessary to determine whether the path collides with an obstacle. If a collision occurs, the path is not added.

[0081] There are many ways to represent distance cost, and this embodiment is not limited to one method. For example, it can be the distance between a node and the destination. The preset threshold is a very small value. If the distance cost between a node and the destination is less than the preset threshold, it means that the extended node is very close to the destination, and the vehicle can obtain a path to park in the target parking space from the node.

[0082] Before the first growth of the random expansion tree, the random expansion tree already has multiple base nodes. Therefore, the growth process of the random expansion tree is accelerated, and the collision detection that is frequently triggered during the growth of the random expansion tree in narrow parking spaces is avoided, which would lead to invalid expansion and greatly improve the search speed.

[0083] In some embodiments, the distance cost between any node of the randomly expanded tree and the endpoint can be determined, but is not limited to, in the following manner:

[0084] If the vehicle moves to any node in the randomly expanded tree but has not entered the driving lane, then the second heading angle of the vehicle at that node, the first distance between the vehicle and the centerline of the driving lane, and the tilt angle of the centerline of the driving lane are obtained. Based on the second heading angle, the tilt angle of the centerline of the driving lane, and the first distance, the distance cost between that node and the destination is determined. distance It can be expressed as the following formula (3);

[0085] cost distance =k1·dist(c,l r ) 2 +k2·(θ l -θ c2 ) 2 (3)

[0086] In formula (3), k1 and k2 are preset weight parameters, dist(c,l) r ) represents the first distance between the vehicle and the center line of the driving lane, c represents the center of the rear axle of the vehicle, and l r Indicates the center line of the driving lane, θ l θ is the inclination angle of the lane centerline (the lane centerline is usually horizontal). c2 This is the second heading angle;

[0087] If the vehicle has entered the driving lane when it moves to any node in the randomly expanded tree, then obtain the third heading angle and second position coordinates of the vehicle when it is at that node, and the destination heading angle and destination position coordinates of the vehicle when it is at the destination. Then, determine the distance cost between the arbitrary node and the destination based on the third heading angle, second position coordinates, destination heading angle, and destination position coordinates. distance It can be expressed as the following formula (4);

[0088] cost distance =k3·(x g -x c ) 2 +k4·(y g -y c ) 2 +k5·(θ g -θ c3 ) 2 (4)

[0089] In formula (4), k3, k4, and k5 are preset weight parameters, and θ g θ is the heading angle at the destination.c3 The third heading angle; the second position coordinates include the lateral coordinate x of the vehicle's rear axle center. c and vertical coordinate y c The endpoint coordinates include the horizontal coordinate x. g and vertical coordinate y g .

[0090] Specifically, for path planning in parking scenarios, planning within the driving lane is relatively easy. The difficulty and key point lie in the path planning from the parking space to the driving lane. Therefore, different search strategies should be used before and after the vehicle enters the driving lane from the parking space. Entering the driving lane is defined as when the vehicle's outline has completely left the parking space. When the vehicle is driving in the driving lane and planning begins, it is mostly in a nearly horizontal state, that is, the vehicle's direction is parallel to the driving lane and perpendicular to the parking space. Therefore, the planning goal before entering the driving lane from the parking space is set as coinciding with the center line of the driving lane, focusing on the vehicle's longitudinal coordinate and heading angle, but not the lateral coordinate. The planning goal after entering the driving lane is the planning endpoint, that is, the vehicle's current position, at which point all attitude parameters should be considered.

[0091] It should be noted that this embodiment uses a variable distance cost, incorporating the attraction of the endpoint and the center line of the driving lane on the growth direction of the expansion tree, thereby reducing the time cost increased by the expansion tree due to random and aimless growth.

[0092] In some embodiments, each growth of the randomly expanding tree includes:

[0093] Step S301: Random sampling is performed to obtain temporary sampling points, and a node is selected from the existing nodes of the random expansion tree as the nearest neighbor node of the temporary sampling point. Specifically, the nearest neighbor node can be the node closest to the temporary sampling point, and the range of random sampling can be the target parking space plus a part of the driving lane adjacent to the parking space. The range can be appropriately expanded outward or contracted inward from the main body according to the actual debugging situation.

[0094] Step S302: From the neighboring node, move a preset step length in multiple directions according to multiple preset wheel turning angles, and perform collision prediction based on the obstacle information; specifically, multiple initial child nodes can be obtained by moving in multiple directions, and then collision prediction is performed on these multiple points. The principle of collision prediction is described in the previous text and will not be repeated here.

[0095] Step S303: If no collision occurs after moving in any direction as in step S302, the position after moving in any direction (initial child node) is taken as a candidate child node; a candidate child node indicates that it can be used as a new node.

[0096] In step S304, if there is at least one candidate child node, select the candidate child node with the smallest distance cost to the endpoint from the at least one candidate child node, and use it as a new node in the random expansion tree. Its parent node is the nearest neighbor node, thus completing one growth cycle. Specifically, if there is no candidate child node, it means that no suitable new node has been found, and the next growth cycle needs to be performed again. Steps S301 to S304 are still executed. After multiple growth cycles, a node is finally found whose distance cost to the endpoint is less than a preset threshold, thus completing the growth of the random expansion tree.

[0097] In some embodiments, selecting a candidate child node from the at least one candidate child node that has the minimum distance cost to the endpoint specifically includes:

[0098] If the vehicle is located at any candidate sub-node but has not entered the driving lane, then obtain the fourth heading angle of the vehicle at that candidate sub-node, the second distance between the vehicle and the centerline of the driving lane, and the tilt angle of the centerline of the driving lane. Based on the fourth heading angle, the second distance, and the destination heading angle, determine the distance cost between that candidate sub-node and the destination. distance It can be expressed as the following formula (5);

[0099] cost distance =k1·dist(c,l r ) 2 +k2·(θ l -θ c4 ) 2 (5)

[0100] In formula (3), k1 and k2 are preset weight parameters, dist(c,l) r ) represents the first distance between the vehicle and the center line of the driving lane, c represents the center of the rear axle of the vehicle, and l r Indicates the center line of the driving lane, θ l θ is the inclination angle of the lane centerline (the lane centerline is usually horizontal). c4 This is the second heading angle;

[0101] If the vehicle has entered the driving lane when it is located at any candidate sub-node, then obtain the fifth heading angle and third position coordinates when the vehicle is located at that candidate sub-node, and the destination heading angle and destination position coordinates when the vehicle is located at the destination. Based on the fifth heading angle, third position coordinates, destination heading angle, and destination position coordinates, determine the distance cost between that candidate sub-node and the destination.

[0102] cost distance It can be expressed as the following formula (4);

[0103]

[0104] In formula (4), k3, k4, and k5 are preset weight parameters, and θ g θ is the heading angle at the destination. c5 The fifth heading angle; the third position coordinates include the lateral coordinate x of the vehicle's rear axle center. c and vertical coordinate y c The endpoint coordinates include the horizontal coordinate x. g and vertical coordinate y g .

[0105] Specifically, for path planning in parking scenarios, planning within the driving lane is relatively easy. The difficulty and key point lie in the path planning from the parking space to the driving lane. Therefore, different search strategies should be used before and after the vehicle enters the driving lane from the parking space. Entering the driving lane is defined as when the vehicle's outline has completely left the parking space. When the vehicle is driving in the driving lane and planning begins, it is mostly in a nearly horizontal state, that is, the vehicle's direction is parallel to the driving lane and perpendicular to the parking space. Therefore, the planning goal before entering the driving lane from the parking space is set as coinciding with the center line of the driving lane, focusing on the vehicle's longitudinal coordinate and heading angle, but not the lateral coordinate. The planning goal after entering the driving lane is the planning endpoint, that is, the vehicle's current position, at which point all attitude parameters should be considered.

[0106] It should be noted that this embodiment uses a variable distance cost, incorporating the attraction of the endpoint and the center line of the driving lane on the growth direction of the expansion tree, thereby reducing the time cost increased by the expansion tree due to random and aimless growth.

[0107] In some embodiments, to balance search efficiency and search success rate, the multiple preset wheel angles when searching for new nodes based on neighboring nodes may be, but are not limited to, the following:

[0108] φ max , -φ max ;

[0109] Where, φ max The preset maximum wheel turning angle of the vehicle is defined as follows: if the left is the positive direction, when the steering wheel is turned to the left, the wheel turns to the left, and the turning angle is positive. Conversely, when the steering wheel is turned to the right, the wheel turns to the right, and the turning angle is negative.

[0110] Another embodiment of this application provides a parking path planning device, including a module for performing the parking path planning method described in the above embodiments;

[0111] like Figure 5As shown, the module includes:

[0112] Information acquisition module 1 is used to acquire surrounding environmental information from the target parking space to the current position of the vehicle, and to obtain obstacle information, start point and end point based on the surrounding environmental information; wherein, the start point is the position of the target parking space and the end point is the current position of the vehicle;

[0113] The first path search module 2 is used to perform a first path search based on a preset vehicle kinematics model, according to the obstacle information, the starting point and the ending point, to obtain an initial extended tree; wherein, the initial extended tree includes multiple nodes, and the node information of each node includes the position coordinates and heading angle corresponding to the node;

[0114] The second path search module 3 is used to perform a second path search based on the initial expanded tree, according to the obstacle information, the start and end points, and a preset RRT algorithm, to obtain a random expanded tree; wherein, the random expanded tree includes multiple nodes, and the node information of each node includes the position coordinates and heading angle corresponding to the node; and

[0115] The path generation module 4 is used to obtain the parking planning path based on the random expansion tree.

[0116] The parking path planning device described in the above embodiments is merely illustrative. The modules described as separate components may or may not be physically separate. The components of a module may or may not be physical modules; that is, they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected according to actual needs to achieve the purpose of the parking path planning device scheme in the embodiments.

[0117] It should be noted that the parking path planning device of the above embodiments corresponds to the parking path planning method of the above embodiments. Therefore, the parts of the parking path planning device of the above embodiments that are not described in detail can be obtained by referring to the content of the parking path planning method of the above embodiments. That is, the specific steps recorded in the parking path planning method of the above embodiments can be understood as the functions that the parking path planning device of the above embodiments can achieve, and will not be repeated here.

[0118] Furthermore, if the parking path planning device of the above embodiments is implemented as a software functional module and sold or used as an independent product, it can be stored in a computer-readable storage medium.

[0119] Another embodiment of this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the parking path planning method as described in the above embodiments.

[0120] Specifically, the computer-readable storage medium may include any entity or recording medium capable of carrying the computer program instructions, such as a USB flash drive, portable hard drive, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media.

[0121] Another embodiment of this application provides an electronic device including a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the parking path planning method described in the above embodiments.

[0122] The electronic device may also include a bus connecting different components, including memory and processor. The memory may include a computer-readable medium in the form of volatile memory, such as random access memory (RAM) and / or cache memory. The memory may also include at least one program product having a set (e.g., at least one) of program modules configured to perform the functions of the embodiments of this application. The electronic device may also communicate with one or more external devices (e.g., keyboard, pointing device, display, etc.), and with one or more devices that enable a user to interact with the electronic device, and / or with any device (e.g., a network interface card) that enables the electronic device to communicate with one or more other computing devices, such communication may be performed via an input / output (I / O) interface, and the electronic device may also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN), and / or public networks, such as the Internet) via a network adapter.

[0123] The various embodiments of this application have been described above. These descriptions are exemplary and not exhaustive, nor are they limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen to best explain the principles, practical applications, or technological improvements to the embodiments in the market, or to enable others skilled in the art to understand the embodiments disclosed herein.

Claims

1. A parking path planning method characterized by comprising: The method comprises the following steps: obtaining surrounding environment information from a target parking space to a current position of a vehicle, and obtaining obstacle information, a starting point and an ending point according to the surrounding environment information; the starting point is a position of the target parking space, and the ending point is the current position of the vehicle; based on a preset vehicle kinematics model, performing first path searching according to the obstacle information, the starting point and the ending point to obtain an initial expansion tree; the initial expansion tree comprises a plurality of nodes, and node information of each node comprises position coordinates and a heading angle corresponding to the node; based on the initial expansion tree, performing second path searching according to the obstacle information, the starting point and the ending point and a preset RRT algorithm to obtain a random expansion tree; the random expansion tree comprises a plurality of nodes, and node information of each node comprises position coordinates and a heading angle corresponding to the node; obtaining a parking planning path according to the random expansion tree.

2. The parking route planning method according to claim 1, characterized by, The first path searching according to the starting point and the ending point specifically comprises the following steps: taking the starting point as a root node of the initial expansion tree to perform node expansion in a plurality of expansion directions, and each expansion direction corresponds to a preset wheel rotation angle; when performing node expansion in any expansion direction, the vehicle moves a preset step length from a current node according to the preset wheel rotation angle, and performs collision prediction according to the obstacle information; if no collision occurs after the movement, a position after the movement is taken as a new node in the expansion direction, and whether the node expansion in the expansion direction is ended is judged; if collision occurs after the movement, the vehicle moves a preset step length directly from the current node, and a position after the movement is taken as a new node in the expansion direction, and whether the node expansion in the expansion direction is ended is judged; if the position of the new node meets a preset condition, the expansion is ended, and if the position of the new node does not meet the preset condition, the node expansion in the expansion direction is continued with the new node as the current node.

3. The parking route planning method according to claim 2, characterized by, The first path searching according to the starting point and the ending point further comprises the following steps: when a new node is obtained, a first position coordinate and a first heading angle corresponding to a parent node of the new node, a first wheel rotation angle and a moving speed of the vehicle from the parent node to the new node are obtained, and a position coordinate and a heading angle of the new node are calculated according to the first position coordinate, the first heading angle, the first wheel rotation angle and the moving speed based on the preset vehicle kinematics model.

4. The parking route planning method according to claim 2, characterized by, The preset condition comprises the following steps: when the vehicle moves to the new node, the vehicle meets a condition of driving out of the target parking space; the condition of driving out of the target parking space is that a vehicle body completely drives out of a parking line of the target parking space, or a vehicle head touches a road boundary above the target parking space.

5. The parking route planning method according to claim 2, characterized by, The plurality of preset wheel rotation angles corresponding to the plurality of expansion directions are respectively: wherein φ max is a preset maximum vehicle wheel angle.

6. The parking route planning method according to claim 1, characterized by, The second path searching based on the initial expansion tree, the obstacle information, the starting point and the ending point and the preset RRT algorithm to obtain the random expansion tree specifically comprises the following steps: The multiple nodes of the initial expansion tree are taken as multiple base nodes of the random expansion tree, and at least one growth of the random expansion tree is performed based on the multiple base nodes until the distance cost between any node of the random expansion tree and the terminal point is less than a preset threshold, and then the growth is stopped, and the second path search is ended.

7. The parking route planning method according to claim 6, characterized by, The distance cost between any node of the random expansion tree and the terminal point is determined according to the following manner: If the vehicle has entered the lane when the vehicle moves to any node of the random expansion tree, the fourth heading angle and the second distance between the vehicle and the lane center line, the lane center line inclination angle when the vehicle is located at the any node are obtained, and the distance cost between the any node and the terminal point is determined according to the fourth heading angle, the second distance, the terminal heading angle and the terminal position coordinates. If the vehicle has entered the lane when the vehicle moves to any node of the random expansion tree, the fourth heading angle and the second distance between the vehicle and the lane center line, the lane center line inclination angle when the vehicle is located at the any node are obtained, and the distance cost between the any node and the terminal point is determined according to the fourth heading angle, the second distance, the terminal heading angle and the terminal position coordinates.

8. The parking route planning method according to claim 6, characterized by, Each growth of the random expansion tree includes: A temporary sampling point is obtained by random sampling, and a node of the random expansion tree is selected as a near neighbor node of the temporary sampling point; From the near neighbor node, a preset step is moved in multiple directions according to multiple preset wheel rotation angles, and collision prediction is performed according to the obstacle information; If no collision occurs after moving in any direction, the position after moving in any direction is taken as a candidate child node; If there is at least one candidate child node, a candidate child node with the minimum distance cost between the candidate child node and the terminal point is selected from the at least one candidate child node as a new node of the random expansion tree, and the parent node of the new node is the near neighbor node, and thus one growth is completed.

9. The parking route planning method according to claim 8, characterized by, The selection of the candidate child node with the minimum distance cost between the candidate child node and the terminal point includes: If the vehicle has not entered the lane when the vehicle is located at any candidate child node, the fourth heading angle and the second distance between the vehicle and the lane center line, the lane center line inclination angle when the vehicle is located at the any candidate child node are obtained, and the distance cost between the any candidate child node and the terminal point is determined according to the fourth heading angle, the second distance, the terminal heading angle and the terminal position coordinates. If the vehicle has entered the lane when the vehicle is located at any candidate child node, the fifth heading angle and the third position coordinates when the vehicle is located at the any candidate child node, the terminal heading angle and the terminal position coordinates when the vehicle is located at the terminal point are obtained, and the distance cost between the any candidate child node and the terminal point is determined according to the fifth heading angle, the third position coordinates, the terminal heading angle and the terminal position coordinates.

10. The parking route planning method according to claim 8, characterized by, The multiple preset wheel rotation angles include: wherein φ max is a preset maximum vehicle wheel angle.

11. A parking path planning apparatus characterized by comprising: The module for performing the parking path planning method of any one of claims 1-10. The module for performing the parking path planning method of any one of claims 1-10.

12. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a computer program, and the computer program, when executed by a processor, implements the parking path planning method in any one of claims 1-10.

13. An electronic device comprising a processor, a memory, and a computer program stored on the memory and executable on the processor, characterized in that, The processor executes the computer program to implement the parking path planning method in any one of claims 1-10.

Images