Driving path planning method, device, storage medium and electronic device

By dividing obstacle grid groups in the path planning of autonomous driving vehicles and creating a search tree for circular areas, combining cost estimation function and path planning algorithm, the problems of low efficiency and poor effect of path planning in the existing technology are solved, and the problem of low path planning efficiency and poor effect is realized. Shorter and obstacle-avoiding path planning.

CN118518135BActive Publication Date: 2024-11-15BYD CO LTD +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410996319.8
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-07-24
Publication Date
2024-11-15
Estimated Expiration
2044-07-24

AI Technical Summary

Technical Problem

The existing path planning methods for autonomous driving vehicles cannot efficiently plan paths with short paths and obstacle avoidance, and the path planning effect is poor and the efficiency is low.

Method used

By obtaining the grid map of the driving environment, dividing the obstacle grid into multiple grid groups, and creating a search tree for circular areas for each group, planning using the cost estimation function based on the vehicle's origin and destination locations Driving paths are combined with RS curve method and A* planning algorithm to alternately plan paths.

Benefits of technology

It improves the speed and effect of path planning, reduces the probability of path collision with obstacles, and generates a shorter path and can effectively avoid obstacles.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118518135B_ABST
    Figure CN118518135B_ABST
Patent Text Reader

Abstract

The present application discloses a driving path planning method, device, storage medium and electronic device, belonging to the field of automatic driving technology. The driving path planning method comprises: obtaining a grid map of a driving environment and obstacle grids on the grid map; dividing all obstacle grids into a plurality of obstacle grid groups; creating a search tree corresponding to each obstacle grid group, wherein each tree node in the search tree corresponds to a circular area; determining the driving path of the mobile vehicle based on the starting position, destination position and search tree of the mobile vehicle, thereby efficiently planning a short path that can avoid obstacles, effectively reducing the collision probability between the planned path and obstacles, improving the path planning speed and achieving a good path planning effect.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application belongs to the field of autonomous driving technology, and in particular, relates to a driving path planning method, device, storage medium, electronic device and computer program product. Background Art

[0002] With the development of the automobile industry and smart technology, car owners have higher and higher requirements for the intelligence and convenience of vehicles. For example, given a destination, a smart vehicle can plan a path to avoid obstacles and automatically drive to the destination. This function is applicable to scenarios such as smart warehouses, smart ports, and smart mines, and is also applicable to drivers summoning vehicles.

[0003] There are many algorithms for path planning of autonomous vehicles, such as the geometric method of splicing straight lines and arcs and the grid occupancy method. The planned route needs to ensure that there is no collision between the vehicle and obstacles and a safe distance is left. However, the existing path planning methods cannot efficiently plan a short path that can avoid obstacles, and the path planning effect is poor and the efficiency is low. Summary of the invention

[0004] The present application aims to solve at least one of the technical problems existing in the prior art. To this end, the present application proposes a driving path planning method, device, storage medium, electronic device and computer program product, which can efficiently plan a short path that can avoid obstacles and has a good path planning effect.

[0005] In a first aspect, the present application provides a driving path planning method, comprising:

[0006] Obtaining a grid map of a driving environment and an obstacle grid on the grid map;

[0007] Dividing all the obstacle grids into a plurality of obstacle grid groups;

[0008] Creating a retrieval tree corresponding to each obstacle grid group, wherein each tree node in the retrieval tree corresponds to a circular area;

[0009] A driving path of the mobile vehicle is determined based on the originating position of the mobile vehicle, the destination position and the search tree.

[0010] In some embodiments, dividing all the obstacle grids into a plurality of obstacle grid groups includes:

[0011] Performing principal component analysis on all the obstacle grids to determine the principal component direction, and taking the direction perpendicular to the principal component direction as the extension direction of the segmentation line to be determined;

[0012] Determining a segmentation line according to the principal component direction, the extension direction and the obstacle grid;

[0013] All the obstacle grids are divided into a plurality of obstacle grid groups according to the dividing lines.

[0014] In some embodiments, the determining the segmentation line according to the principal component direction, the extension direction and the obstacle grid includes:

[0015] Determining an intersection point between the to-be-determined segmentation line and the principal component direction according to the obstacle grid;

[0016] A straight line passing through the intersection point and extending along the extending direction is determined as a dividing line.

[0017] In some embodiments, the creating a search tree corresponding to each obstacle grid group includes:

[0018] Determine a minimum circle that includes all the obstacle grids in the obstacle grid group;

[0019] From the obstacle grid group, select the obstacle grid farthest from the center of the minimum circle as a first reference grid, and select the obstacle grid farthest from the first reference grid as a second reference grid;

[0020] A corresponding retrieval tree is created according to the first reference grid, the second reference grid and the obstacle grid group.

[0021] In some embodiments, creating a corresponding search tree according to the first reference grid, the second reference grid and the obstacle grid group includes:

[0022] Determine a first distance between each obstacle grid in the obstacle grid group and the first reference grid, and a second distance between each obstacle grid and the second reference grid;

[0023] Dividing the obstacle grid group into two subgroups according to the first distance and the second distance;

[0024] A corresponding search tree is created according to the subgroup, the smallest circle including all the obstacle grids in the subgroup, the obstacle grid group, and the smallest circle including all the obstacle grids in the obstacle grid group.

[0025] In some embodiments, determining the driving path of the mobile vehicle based on the starting position, the destination position and the search tree of the mobile vehicle includes:

[0026] According to the starting position, destination position and the retrieval tree of the mobile vehicle, a driving path of the mobile vehicle is planned on the grid map through a cost estimation function, wherein an input item of the cost estimation function includes the shortest distance between a path node and a nearest obstacle.

[0027] In some embodiments, planning the driving path of the mobile vehicle on the grid map by using a cost estimation function according to the starting position, the destination position and the search tree of the mobile vehicle includes:

[0028] Determining a current planned driving path, the initially planned driving path being a starting position of the moving vehicle;

[0029] Taking the path node at the end of the planned driving path as the current starting point of the mobile vehicle;

[0030] Determining a corresponding first driving path on the grid map by using an RS curve method according to the current starting point and the destination position;

[0031] When the first driving path does not collide with an obstacle, the planned driving path and the first driving path are spliced ​​to obtain a final driving path of the mobile vehicle;

[0032] When the first driving path collides with the obstacle, a second driving path of the mobile vehicle is planned on the grid map through a cost estimation function according to the current starting point, the destination position and the retrieval tree, and a final driving path of the mobile vehicle is determined according to the planned driving path and the second driving path.

[0033] In some embodiments, planning the second driving path of the mobile vehicle on the grid map by a cost estimation function according to the current starting point, the destination location and the search tree includes:

[0034] When planning a current path node in the second driving path on the grid map, determining a plurality of current candidate grids;

[0035] Determine the shortest distance between the current candidate grid and the nearest obstacle according to the search tree;

[0036] Determine data of input items in a cost estimation function according to the shortest distance, the current candidate grid, the previous path node, the current starting point position and the destination position, and substitute the data into the input items to perform cost calculation;

[0037] A current path node is determined from the current candidate grid according to the calculated cost.

[0038] In some embodiments, determining the shortest distance between the current candidate grid and the nearest obstacle according to the search tree includes:

[0039] Determining the required safety circle diameter of the mobile vehicle during driving;

[0040] Determine a target circle with the position of the current candidate grid as the center and the diameter of the safety circle as the diameter;

[0041] Determine, from the search tree, a leaf node intersecting with the target circle according to the center and radius corresponding to each node in the search tree;

[0042] The minimum Euclidean distance between the current candidate grid and the obstacle grid in the intersecting leaf node is calculated to obtain the shortest distance between the current candidate grid and the nearest obstacle.

[0043] In a second aspect, the present application provides a driving path planning device, comprising:

[0044] An acquisition unit, used to acquire a grid map of the driving environment and obstacle grids on the grid map;

[0045] A division unit, used for dividing all the obstacle grids into a plurality of obstacle grid groups;

[0046] A creating unit, used for creating a retrieval tree corresponding to each obstacle grid group, wherein each tree node in the retrieval tree corresponds to a circular area;

[0047] A planning unit is used to determine a driving path of the mobile vehicle based on a starting position, a destination position and the retrieval tree of the mobile vehicle.

[0048] In some embodiments, the dividing unit is specifically used for:

[0049] Performing principal component analysis on all the obstacle grids to determine the principal component direction, and taking the direction perpendicular to the principal component direction as the extension direction of the segmentation line to be determined;

[0050] Determining a segmentation line according to the principal component direction, the extension direction and the obstacle grid;

[0051] All the obstacle grids are divided into a plurality of obstacle grid groups according to the dividing lines.

[0052] In some embodiments, the dividing unit is specifically used for:

[0053] Determining an intersection point between the to-be-determined segmentation line and the principal component direction according to the obstacle grid;

[0054] A straight line passing through the intersection point and extending along the extending direction is determined as a dividing line.

[0055] In some embodiments, the creating unit is specifically used to:

[0056] Determine a minimum circle that includes all the obstacle grids in the obstacle grid group;

[0057] From the obstacle grid group, select the obstacle grid farthest from the center of the minimum circle as a first reference grid, and select the obstacle grid farthest from the first reference grid as a second reference grid;

[0058] A corresponding retrieval tree is created according to the first reference grid, the second reference grid and the obstacle grid group.

[0059] In some embodiments, the creating unit is specifically used to:

[0060] Determine a first distance between each obstacle grid in the obstacle grid group and the first reference grid, and a second distance between each obstacle grid and the second reference grid;

[0061] Dividing the obstacle grid group into two subgroups according to the first distance and the second distance;

[0062] A corresponding search tree is created according to the subgroup, the smallest circle including all the obstacle grids in the subgroup, the obstacle grid group, and the smallest circle including all the obstacle grids in the obstacle grid group.

[0063] In some embodiments, the planning unit is specifically used to:

[0064] According to the starting position, destination position and the retrieval tree of the mobile vehicle, a driving path of the mobile vehicle is planned on the grid map through a cost estimation function, wherein an input item of the cost estimation function includes the shortest distance between a path node and a nearest obstacle.

[0065] In some embodiments, the planning unit is specifically used to:

[0066] Determining a current planned driving path, the initially planned driving path being a starting position of the moving vehicle;

[0067] Taking the path node at the end of the planned driving path as the current starting point of the mobile vehicle;

[0068] Determining a corresponding first driving path on the grid map by using an RS curve method according to the current starting point and the destination position;

[0069] When the first driving path does not collide with an obstacle, the planned driving path and the first driving path are spliced ​​to obtain a final driving path of the mobile vehicle;

[0070] When the first driving path collides with the obstacle, a second driving path of the mobile vehicle is planned on the grid map through a cost estimation function according to the current starting point, the destination position and the retrieval tree, and a final driving path of the mobile vehicle is determined according to the planned driving path and the second driving path.

[0071] In some embodiments, the planning unit is specifically used to:

[0072] When planning a current path node in the second driving path on the grid map, determining a plurality of current candidate grids;

[0073] Determine the shortest distance between the current candidate grid and the nearest obstacle according to the search tree;

[0074] Determine data of input items in a cost estimation function according to the shortest distance, the current candidate grid, the previous path node, the current starting point position and the destination position, and substitute the data into the input items to perform cost calculation;

[0075] A current path node is determined from the current candidate grid according to the calculated cost.

[0076] In some embodiments, the planning unit is specifically used to:

[0077] Determining the required safety circle diameter of the mobile vehicle during driving;

[0078] Determine a target circle with the position of the current candidate grid as the center and the diameter of the safety circle as the diameter;

[0079] Determine, from the search tree, a leaf node intersecting with the target circle according to the center and radius corresponding to each node in the search tree;

[0080] The minimum Euclidean distance between the current candidate grid and the obstacle grid in the intersecting leaf node is calculated to obtain the shortest distance between the current candidate grid and the nearest obstacle.

[0081] In a third aspect, the present application provides a non-transitory computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the driving path planning method described in any one of the above items is implemented.

[0082] In a fourth aspect, the present application provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements any of the above-described driving path planning methods when executing the computer program.

[0083] In a fifth aspect, the present application provides a computer program product, including a computer program, which, when executed by a processor, implements any of the driving path planning methods described above.

[0084] The driving path planning method, device, storage medium, electronic device and computer program product provided by the embodiments of the present application obtain a grid map of the driving environment and obstacle grids on the grid map; divide all obstacle grids into multiple obstacle grid groups; create a search tree corresponding to each obstacle grid group, and each tree node in the search tree corresponds to a circular area; determine the driving path of the mobile vehicle based on the starting position, destination position and search tree of the mobile vehicle, that is, by dividing the obstacle grid into multiple parts, and creating a search tree for each part that uses circular areas to divide the search area, so as to improve the search speed of the nearest obstacle in the path planning process, so as to efficiently plan a short path that can avoid obstacles, effectively reduce the collision probability of the planned path and obstacles, improve the path planning speed, and achieve good path planning effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0085] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the description of the embodiments in conjunction with the following drawings, in which:

[0086] Figure 1 It is a flowchart of a driving path planning method provided in an embodiment of the present application;

[0087] Figure 2 It is a schematic diagram of an XY plane display of a grid map provided in an embodiment of the present application;

[0088] Figure 3 is another flowchart of the driving path planning method provided in an embodiment of the present application;

[0089] Figure 4 It is a schematic diagram of the search area division of the search tree provided in the embodiment of the present application;

[0090] Figure 5 is a schematic diagram showing a search tree in an embodiment of the present application;

[0091] Figure 6 It is a schematic diagram showing a safety circle provided in an embodiment of the present application;

[0092] Figure 7is a schematic diagram of the structure of a driving path planning device provided in an embodiment of the present application;

[0093] Figure 8 is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application;

[0094] Fig. 9 It is a schematic diagram of the hardware structure of the electronic device provided in the embodiment of the present application. DETAILED DESCRIPTION

[0095] The embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present application, and cannot be understood as limiting the present application.

[0096] The technical terms used in this application are as follows:

[0097] Search tree: A search table that is created by dividing the space into specific shapes.

[0098] A* planning algorithm: A rough path planning algorithm based on sampling search proposed in 1968. It is the most effective direct search method for solving the shortest path in a static road network.

[0099] RS curve: The full name is Reeds-Shepp curve, which is a path planning algorithm composed of arcs and straight lines that allows vehicles to move forward and backward. Given a starting point and an end point, the RS curve can calculate the shortest path between two points. The path consists of an arc (circle of maximum curvature) and a straight line segment.

[0100] There are many algorithms for path planning of autonomous vehicles, such as the geometric method of splicing straight lines and arcs (RS curve) and the grid occupancy method. The planned route needs to ensure that there is no collision between the vehicle and obstacles and a safe distance is left. Therefore, when planning the path, if the location information of the obstacle is used as an input parameter, it is conducive to planning a short path that can avoid obstacles. How to determine the nearest obstacle at each path point has become a key point.

[0101] In some embodiments, a KD (K-dimensional tree) tree about obstacles can be established, and the location of the nearest obstacle can be obtained by searching the KD tree. The KD tree is a binary tree structure used to segment and organize data in a multidimensional space. However, on the one hand, since the divided area corresponding to each node in the current KD tree is a rectangle, and the distances from each point on the rectangular boundary to the center point of the rectangle are inconsistent, the location retrieval of the nearest obstacle needs to be retrieved and compared on more branches of the KD tree, which reduces the retrieval speed. On the other hand, since the balance of the spatial distribution of obstacles is not considered when dividing the area corresponding to each node, the established KD tree has a large number of branch layers and a deep tree depth, which is not conducive to improving the retrieval speed of the nearest obstacle, thereby affecting the planning speed of the driving path.

[0102] To solve at least one of the above technical problems, an embodiment of the present application provides a driving path planning method, device, storage medium, electronic device and computer program product.

[0103] See also Figure 1 , Figure 1 1 is a flow chart of a driving path planning method provided in an embodiment of the present application. The driving path planning method is applied to an electronic device, and the electronic device can be implemented as an autonomous driving vehicle. The driving path planning method may include the following steps 101-104, wherein:

[0104] 101. Obtain a grid map of the driving environment and an obstacle grid on the grid map.

[0105] Among them, see Figure 2 , Figure 2 It is an XY plane display schematic diagram of the grid map provided in the embodiment of the present application. The driving environment can be a scene environment such as a parking lot, a dock, a warehouse, a mine, etc. The grid map is a map obtained by gridding the driving environment, which includes multiple cells, each of which represents a location point. The grid map can be automatically sensed and obtained by on-board sensors or field-side sensors (such as lidar), or it can be obtained by other means, such as manual drawing, which is not limited here.

[0106] Obstacle grid (e.g. Figure 2The grids filled with shadows in the figure are grids marked with obstacles, which are mainly static obstacles. Considering that the collision of vehicles first occurs at the outer contour of the obstacle, it is only necessary to mark the position of the outer contour of the obstacle on the grid map, without marking the entire obstacle position, that is, the obstacle grid is the grid corresponding to the position of the outer contour of the obstacle. This marking method will not affect the accuracy of predicting whether an obstacle collision will occur in the subsequent path planning process, and can reduce the number of obstacle grids, and then when creating a retrieval tree based on the obstacle grid, it is helpful to reduce the size of the retrieval tree and improve the retrieval speed.

[0107] 102. Divide all obstacle grids into multiple obstacle grid groups.

[0108] Among them, the obstacles can be analyzed by principal component analysis first to obtain the direction of the principal component, and all obstacles can be divided into two parts (that is, divided into two obstacle grid groups) using a dividing line perpendicular to the direction of the principal component, and then a search tree is established for each part, so that a more balanced search tree with fewer layers can be established. In other implementations, an overall search tree is directly established for all obstacles, and the branches of this search tree are unevenly distributed, have many layers, and the search speed is slow.

[0109] For details, see Figure 3 , Figure 3 1 is another flowchart of the driving path planning method provided in an embodiment of the present application. The above step 102 specifically includes the following steps 1021-1023, wherein:

[0110] 1021. Perform principal component analysis on all obstacle grids to determine the principal component direction, and use the direction perpendicular to the principal component direction as the extension direction of the segmentation line to be determined.

[0111] in, Figure 2 The w direction in is the direction of the principal component vector of the obstacle distribution (principal component direction), and the principal component vector can be obtained using the least squares method.

[0112] For example, suppose the expression of the line where the principal component vector lies is: ;

[0113] make

[0114] (1)

[0115] (2)

[0116] (3)

[0117] in, ~ is the x-axis coordinate value of obstacle 1-n, ~ is the y-axis coordinate value of obstacle 1-n.

[0118] According to the least squares method, we get:

[0119] ;

[0120] Then the principal component vector w is:

[0121] ;

[0122] The unit vectors of the principal components are:

[0123] ;

[0124] At this time, the direction perpendicular to the principal component vector w is the extension direction of the segmentation line to be determined.

[0125] 1022. Determine the segmentation line according to the principal component direction, the extension direction and the obstacle grid.

[0126] In some embodiments, the above step 1022 specifically includes:

[0127] Determine the intersection between the to-be-determined segmentation line and the principal component direction according to the obstacle grid;

[0128] A straight line passing through the intersection point and extending in the extension direction is determined as a dividing line.

[0129] Among them, although the direction of the segmentation line has been determined in step 1021 above, there are countless straight lines corresponding to this direction. In order to specifically determine the position of the segmentation line, it is also necessary to determine the intersection of the segmentation line and the principal component direction (that is, Figure 2 Points in )Location, The determination needs to make the time spent on searching the nearest obstacle in the search tree as short as possible.

[0130] The cost function of the retrieval is designed according to the following principles: Because the retrieval of the nearest obstacle is random, the subtrees of the nodes in the retrieval tree should be as similar as possible, and the fewer the number of layers, the better. Based on the above considerations, the following goals can be obtained:

[0131] ①Only record the data of the outer contour of the obstacle, because the collision starts from the outer contour.

[0132] ② The number of obstacle grids contained in the subtrees under the same node should be as equal as possible.

[0133] ③Reduce the radius of the circle area in the retrieval tree.

[0134] Among them, goal ① has been achieved when marking obstacle grids on the grid map.

[0135] Objective function for goal ②: Assume that there are N grids on the grid map marked as having obstacles. When the number of grids with obstacles in the two subtrees is equal When , goal ② can be fully achieved, and this goal can be transformed into the objective function .

[0136] Objective function for goal ③: When the distance between the two farthest obstacle grids in one subtree is equal to the distance between the two farthest obstacle grids in another subtree, goal ③ is completely achieved. In a two-dimensional coordinate system, the calculation of distance includes squares and square roots. In order to simplify the calculation and improve efficiency, the projection length of the obstacle grid in the principal component direction is used instead of the distance, because the projection is a scalar. For any obstacle grid, the calculation method of its projection length is as follows:

[0137] ,

[0138] when When , the requirements of goal ③ are met, and this goal can be transformed into the objective function ,in, is the intersection point, is the minimum projection length of the obstacle grid in the principal component direction, is the maximum projection length of the obstacle mesh in the principal component direction.

[0139] Considering the above objective functions and , the objective function Get the minimum value to determine the intersection of the dividing line and the principal component direction , that is,

[0140] ;

[0141] ;

[0142] in, yes The weight value can be set manually. For example, if you prefer a balanced number of obstacle grids, The value should be set smaller if you prefer to reduce the radius of the circle in the subtree. The value needs to be set larger.

[0143] 1023. Divide all obstacle grids into multiple obstacle grid groups according to the dividing lines.

[0144] The number of obstacle grid groups may be two. In this case, the obstacle grids on the same side of the dividing line on the grid map are directly divided into one group to obtain two obstacle grid groups.

[0145] That is, please continue to see Figure 2 , the dividing line L is perpendicular to the principal component direction w and passes through point The only straight line is found. At this time, the obstacle grids on the grid map are automatically divided into two groups with the dividing line L as the boundary. The obstacle grids on the same side of the dividing line are classified into the same group. It can be clearly seen from the figure that the number of obstacle grids in the two groups is not much different, so the grouping of obstacle grids is achieved based on the balance of spatial distribution. When creating a retrieval tree for each obstacle grid group, it can ensure that the branches and layers of the two trees are the same or similar, avoiding a large difference in structure between the two (for example, one tree has a larger depth and the other tree has a smaller depth) and affecting the subsequent retrieval speed.

[0146] In other embodiments, the number of obstacle grid groups may be three or more. In this case, the above method is first used to divide all obstacle grids into two groups. The same method can be used to further divide any group, that is, the dividing line is first determined based on the principal component analysis method, and then the dividing line is used as the boundary to divide the group into two groups. The two groups divided this time can still be further divided by the same method.

[0147] 103. Create a retrieval tree corresponding to each obstacle grid group, and each tree node in the retrieval tree corresponds to a circular area.

[0148] Among them, for obstacles in the entire driving environment, two retrieval trees with equal or similar structures can be created, and this solution is to use a circular area as the retrieval area to create a retrieval tree, and each tree node corresponds to a circular area. Since the distance from each point on the circular boundary to the center point is the same, when the nearest obstacle position of the path node is subsequently retrieved in the retrieval tree, the number of branches and regions to be retrieved can be reduced, thereby improving the retrieval speed. In other implementations, a rectangular area is used as the retrieval area to create a retrieval tree, and each tree node corresponds to a rectangular area. Since the distance from each point on the rectangular boundary to the center point is inconsistent, the retrieval of the nearest obstacle position needs to be searched and compared in more branches and more areas of the retrieval tree, which reduces the retrieval speed.

[0149] In some embodiments, see Figure 3 , the above step 103 may include the following steps 1031-1033, wherein:

[0150] 1031. Determine the smallest circle that contains all obstacle grids in the obstacle grid group.

[0151] The smallest circle refers to the circle with the smallest radius among all circles that contain all obstacle meshes in the obstacle mesh group. Figure 4 , Figure 4 : This is a schematic diagram of the search area division of the search tree provided in the embodiment of the present application. Each circular area is a search area. The circle h1 shown in the figure is the smallest circle including all obstacles below the dividing line (i.e., one obstacle grid group mentioned above), and the circle h2 is the smallest circle including all obstacles above the dividing line (i.e., another obstacle grid group mentioned above). For the search area of ​​the search tree corresponding to the obstacles above the dividing line, only the circle h2 is shown. Figure 4 The circles except h2 are the search areas of the search tree corresponding to the obstacles below the dividing line.

[0152] 1032. From the obstacle grid group, select the obstacle grid farthest from the center of the minimum circle as the first reference grid, and select the obstacle grid farthest from the first reference grid as the second reference grid.

[0153] 1033. Create a corresponding retrieval tree based on the first reference grid, the second reference grid and the obstacle grid group.

[0154] The tree nodes in the search tree are arranged layer by layer from top to bottom. These tree nodes include the root node at the top level, at least one layer of child nodes below the root node, and the leaf nodes at the last level. When creating a search tree, it is necessary to first determine the search area (circular area) corresponding to each tree node and the information of the obstacle grid it includes, that is, each tree node contains the following information: the radius of the corresponding circular area, the coordinates of the center of the corresponding circular area, and the information of the obstacle grid it includes.

[0155] Please continue to see Figure 4 and Figure 5 , Figure 5 1 is a schematic diagram showing a search tree in an embodiment of the present application. Taking the area division of the search tree corresponding to the obstacle below the dividing line as an example, the search tree includes 4 layers, and the tree node in the first layer is the root node BT 11 , the second-level tree nodes include BT 21 and BT 22 , the third-level tree nodes include BT 31 ~ BT 34 , the fourth layer of tree nodes are leaf nodes, including BT 41 ~ BT 48 Among them, the root node BT 11 The corresponding search area is circle h1, the root node BT 11 The information to be included includes: the radius of circle h1, the coordinates of the center of circle h1, and the obstacle grid located in circle h1 and below the dividing line.11 The corresponding obstacle grid continues to divide the next layer of two sub-nodes BT 21 and BT 22 Specifically, the search area is as follows: first find the obstacle grid farthest from the center of circle h1 as the first reference grid, and select the obstacle grid farthest from the first reference grid as the second reference grid. The first reference grid and the second reference grid are usually located in different subnodes (BT 21 or BT 22 ) in the search area, based on these two reference grids, the remaining obstacle grids are divided into child nodes BT 21 or BT 22 In the search area, the search area is finally determined based on the obstacle grids corresponding to the two child nodes.

[0156] In some embodiments, the above step 1033 specifically includes:

[0157] Determine a first distance between each obstacle grid in the obstacle grid group and the first reference grid, and a second distance between each obstacle grid and the second reference grid;

[0158] Dividing the obstacle grid group into two subgroups according to the first distance and the second distance;

[0159] A corresponding search tree is created according to the subgroup, the smallest circle including all the obstacle grids in the subgroup, the obstacle grid group, and the smallest circle including all the obstacle grids in the obstacle grid group.

[0160] Among them, the remaining obstacle grids are mainly distributed by calculating the distances between the remaining obstacle grids and the first reference grid and the second reference grid (that is, the first distance and the second distance mentioned above), generally to the one with the closest distance. In this way, two tree nodes of the second layer in the retrieval tree are determined (for example Figure 5 Root node BT 11 The two branch nodes below BT 21 and BT 22 ) and then determine the minimum circle surrounding the corresponding obstacle grid to obtain the search area corresponding to each of the two branch nodes. Then, count the number of obstacle grids corresponding to each branch node and determine whether it exceeds the threshold M. If not, it means that the branch node is a leaf node. If it exceeds, continue to determine the next layer of tree nodes (for example Figure 5 The third level tree node BT 31 ~ BT 34) respectively. At this time, similar to the determination method of the obstacle grid and the retrieval area of ​​the tree node in the current layer, the obstacle grid corresponding to each tree node in the next layer is obtained by further dividing the obstacle grid of the tree node in the current layer into two groups. The division method is also to first determine the first reference grid and the second reference grid, and then assign the remaining obstacle grids to the reference grid that is closest to it. It will not be repeated here.

[0161] After the search area and obstacle grid corresponding to each tree node in the search tree are determined, the search tree can be created, and each tree node stores the radius of its search area, the coordinates of the center of the circle, and the information of the obstacle grid.

[0162] 104. Determine a driving path of a moving vehicle based on an origin position, a destination position, and a search tree of the moving vehicle.

[0163] In some embodiments, please see Figure 3 , the above step 104 may specifically include:

[0164] According to the starting position, the destination position and the retrieval tree of the mobile vehicle, a driving path of the mobile vehicle is planned on the grid map through a cost estimation function, wherein an input item of the cost estimation function includes a closest distance between a path node and a nearest obstacle.

[0165] Among them, the goal of path planning is to find an optimal path from the starting point to the end point on the map, so that the number of cells passed by the path is the least or the path length is the shortest, while avoiding obstacles. According to the selected path planning algorithm, a path search is performed on the map, starting from the starting point, and constantly exploring the surroundings until the end point is found or it is determined that the end point cannot be reached. The nearest obstacle is the obstacle grid closest to the currently searched path node retrieved from the retrieval tree during the path planning process. The closest distance between the path node and the nearest obstacle can also be used as an input item in the heuristic function of the A* planning algorithm, so that when using the A* planning algorithm to plan a path, the probability of collision between the planned path and the obstacle can be effectively reduced.

[0166] In some embodiments, considering that the path planned using the above-mentioned A* planning algorithm can avoid collisions with obstacles, but the planning speed is slow, and the path planned using the conventional RS curve method cannot ensure that there will be no collisions with obstacles, but the speed is relatively fast, the RS curve method and the above-mentioned A* planning algorithm can be combined for path planning.

[0167] Specifically, the above step 104 includes:

[0168] Determine a current planned driving path, where the initial planned driving path is the starting position of the moving vehicle;

[0169] The path node at the end of the planned driving path is used as the current starting point of the moving vehicle;

[0170] Determine a corresponding first driving path on the grid map by using an RS curve method according to the current starting point and the destination position;

[0171] When the first driving path does not collide with the obstacle, the planned driving path and the first driving path are spliced ​​to obtain a final driving path of the mobile vehicle;

[0172] When the first driving path collides with the obstacle, a second driving path of the mobile vehicle is planned on the grid map through a cost estimation function according to the current starting point, the destination position and the retrieval tree, and a final driving path of the mobile vehicle is determined according to the planned driving path and the second driving path.

[0173] That is, when performing path planning, the RS (Reeds-Shepp) curve method and the A* planning algorithm are used alternately for path planning, specifically: ① For the first planning, the RS (Reeds-Shepp) curve method is directly used to plan the path from the vehicle's starting position to the destination position, and it is determined whether the planned path collides with obstacles (that is, whether there are obstacle grids on the planned path). If there is no collision, the path planning is completed. If the planned path collides with obstacles, step ② is executed. ② From the starting position, the A* planning algorithm is used to plan a path of q (a constant set by humans) grids, and step ③ is executed. ③ The end point of the planned path is used as the new heuristic position, and the RS curve method is used to plan the remaining path, and it is determined whether there is a collision with obstacles. If there is no collision, the previously planned path (that is, the path of q grids planned by the A* planning algorithm) and the path planned by the RS curve method are spliced, and smoothing is performed at the splicing point to obtain the final path. The purpose of smoothing is to reduce the inflection points on the path. If there is still a collision, step ④ is executed. ④ Take the end point of the planned path as the new heuristic position and continue to use the A* planning algorithm to plan the path of q grids. If the path has reached the destination during planning, the path planning is completed. If it has not reached the destination, return to step ③.

[0174] Specifically, in step ② or ④, when the A* planning algorithm is used to plan the path of q grids, for any grid n, its cost is calculated by the cost estimation function, and the q grids finally planned must minimize the cost value. That is, the above step "based on the current starting point, the destination location and the search tree, the second driving path of the mobile vehicle is planned on the grid map by the cost estimation function", specifically includes:

[0175] When planning a current path node in a second driving path on the grid map, determining a plurality of current candidate grids;

[0176] Determine the shortest distance between the current candidate grid and the nearest obstacle according to the retrieval tree;

[0177] Determine the data of the input item in the cost estimation function according to the shortest distance, the current candidate grid, the previous path node, the current starting point position and the destination position, and substitute the data into the input item to perform cost calculation;

[0178] The current path node is determined from the current candidate grid according to the calculated cost.

[0179] Among them, the cost estimation function is:

[0180] ;

[0181] ;

[0182] ;

[0183] ;

[0184] ;

[0185] ;

[0186] in, is the cost from the starting location of the path planning here to the current candidate grid (any grid n), is the cost of the current candidate grid, is the cost of the parent grid of the current candidate grid (that is, the previous grid planned in this path planning), is the Euclidean distance between the current candidate grid and the parent grid. It is the Euclidean distance from the current candidate grid to the grid where the destination location is located. It is the difference between the body yaw angle in the current candidate grid path and the body yaw angle of the parent grid. is the shortest distance between the current candidate grid and the nearest obstacle. Is the penalty value representing whether the driving direction needs to be switched. , , , , , u are both constants.

[0187] In some embodiments, the nearest obstacle of the current candidate grid is retrieved in a search tree, and the above step of "determining the shortest distance between the current candidate grid and the nearest obstacle according to the search tree" specifically includes:

[0188] Determine the diameter of the safety circle required for the mobile vehicle during driving;

[0189] The target circle is determined with the position of the current candidate grid as the center and the diameter of the safety circle as the diameter;

[0190] According to the center and radius of each node in the search tree, a leaf node intersecting with the target circle is determined from the search tree;

[0191] The minimum Euclidean distance between the current candidate grid and the obstacle grid in the intersecting leaf node is calculated to obtain the shortest distance between the current candidate grid and the nearest obstacle.

[0192] That is, you need to first determine the safety circle of the moving vehicle. The safety circle is the safe space in which the vehicle does not collide with other objects during driving. Figure 6 , Figure 6 is a schematic diagram showing the safety circle provided in the embodiment of the present application. Specifically, the vehicle is first projected onto the ground to obtain a two-dimensional image of the vehicle, and the geometric center point of the vehicle is set as , the diagonal length is , established with As the center of the circle, The circle with the diameter of is the safety circle required by the vehicle. The diameter of the safety circle is , d is an empirical value, such as 1 meter. Then, the current candidate grid in the path search process is determined in the search tree The nearest obstacle is found, and then the distance between the nearest obstacle and the current candidate grid is calculated to get the nearest distance.

[0193] Among them, when determining the nearest obstacle, the current candidate grid As the center of the circle, with the safety circle diameter ( ) is used as the diameter to determine a target circle, and for the two retrieval trees about the obstacle location information created in the above step 103, all leaf nodes intersecting with the target circle are searched based on the center coordinates and radius of each tree node, and then the nearest obstacle is determined from the obstacle grids corresponding to these leaf nodes.

[0194] When searching for leaf nodes that intersect with the target circle, it is necessary to search for intersecting nodes in each tree node in the retrieval tree from top to bottom. Only when the intersecting tree node is found will the search continue until the leaf node is found. To the center of a tree node in the retrieval tree The distance is used to determine whether they intersect, and the judgment criteria are:

[0195] like , then the target circle does not intersect with the tree node;

[0196] like , then the target circle intersects with the tree node.

[0197] in, is the circle radius of the corresponding tree node.

[0198] If it does not intersect with a certain tree node, stop searching the branch of the tree node. If it is detected that it intersects with the tree node, continue searching the child nodes of the tree node until the leaf node. In this way, all intersecting leaf nodes are found. After all intersecting leaf nodes are found, by calculating The nearest obstacle is determined by the Euclidean distance to the obstacle grid in the intersecting leaf node. The nearest obstacle is the obstacle grid corresponding to the smallest Euclidean distance.

[0199] As can be seen from the above, the driving path planning method provided by the embodiment of the present application obtains a grid map of the driving environment and obstacle grids on the grid map; divides all obstacle grids into multiple obstacle grid groups; creates a retrieval tree corresponding to each obstacle grid group, and each tree node in the retrieval tree corresponds to a circular area; based on the starting position, destination position and retrieval tree of the mobile vehicle, the driving path of the mobile vehicle is determined, that is, by dividing the obstacle grid into multiple parts, a retrieval tree that uses circular areas to divide the retrieval area is created for each part separately, so as to improve the search speed of the nearest obstacle during the path planning process, so as to efficiently plan a short path that can avoid obstacles, effectively reduce the collision probability of the planned path and obstacles, improve the path planning speed, and achieve good path planning effect.

[0200] According to the method described in the above embodiment, the present application embodiment also provides a driving path planning device for executing the steps in the above driving path planning method. Figure 7 , Figure 7 200 is a schematic diagram of the structure of a driving path planning device provided in an embodiment of the present application. The driving path planning device 200 is applied to an electronic device, and the electronic device can be implemented as an autonomous driving vehicle. Specifically, the driving path planning device 200 includes an acquisition unit 201, a division unit 202, a creation unit 203 and a planning unit 204, wherein:

[0201] An acquisition unit 201 is used to acquire a grid map of a driving environment and obstacle grids on the grid map;

[0202] A division unit 202, used to divide all obstacle grids into a plurality of obstacle grid groups;

[0203] A creating unit 203, used to create a search tree corresponding to each obstacle grid group, each tree node in the search tree corresponds to a circular area;

[0204] The planning unit 204 is used to determine the driving path of the mobile vehicle based on the starting position, the destination position and the search tree of the mobile vehicle.

[0205] In some embodiments, the dividing unit 202 is specifically used to:

[0206] Perform principal component analysis on all obstacle grids to determine the principal component direction, and use the direction perpendicular to the principal component direction as the extension direction of the segmentation line to be determined;

[0207] Determine the segmentation line according to the principal component direction, the extension direction and the obstacle grid;

[0208] All obstacle grids are divided into multiple obstacle grid groups according to the dividing lines.

[0209] In some embodiments, the dividing unit 202 is specifically used to:

[0210] Determine the intersection between the to-be-determined segmentation line and the principal component direction according to the obstacle grid;

[0211] A straight line passing through the intersection point and extending in the extension direction is determined as a dividing line.

[0212] In some embodiments, the creation unit 203 is specifically used to:

[0213] Determine a minimum circle that includes all the obstacle grids in the obstacle grid group;

[0214] From the obstacle grid group, select the obstacle grid farthest from the center of the minimum circle as the first reference grid, and select the obstacle grid farthest from the first reference grid as the second reference grid;

[0215] A corresponding retrieval tree is created according to the first reference grid, the second reference grid and the obstacle grid group.

[0216] In some embodiments, the creation unit 203 is specifically used to:

[0217] Determine a first distance between each obstacle grid in the obstacle grid group and the first reference grid, and a second distance between each obstacle grid and the second reference grid;

[0218] Dividing the obstacle grid group into two subgroups according to the first distance and the second distance;

[0219] A corresponding search tree is created according to the subgroup, the smallest circle including all the obstacle grids in the subgroup, the obstacle grid group, and the smallest circle including all the obstacle grids in the obstacle grid group.

[0220] In some embodiments, the planning unit 204 is specifically configured to:

[0221] According to the starting position, the destination position and the retrieval tree of the mobile vehicle, a driving path of the mobile vehicle is planned on the grid map through a cost estimation function, wherein an input item of the cost estimation function includes a closest distance between a path node and a nearest obstacle.

[0222] In some embodiments, the planning unit 204 is specifically configured to:

[0223] Determine a current planned driving path, the initial planned driving path being a starting position of the moving vehicle;

[0224] The path node at the end of the planned driving path is used as the current starting point of the moving vehicle;

[0225] Determine a corresponding first driving path on the grid map by using an RS curve method according to the current starting point and the destination position;

[0226] When the first driving path does not collide with the obstacle, the planned driving path and the first driving path are spliced ​​to obtain a final driving path of the mobile vehicle;

[0227] When the first driving path collides with the obstacle, a second driving path of the mobile vehicle is planned on the grid map through a cost estimation function according to the current starting point, the destination position and the retrieval tree, and a final driving path of the mobile vehicle is determined according to the planned driving path and the second driving path.

[0228] In some embodiments, the planning unit 204 is specifically configured to:

[0229] When planning a current path node in a second driving path on the grid map, determining a plurality of current candidate grids;

[0230] Determine the shortest distance between the current candidate grid and the nearest obstacle according to the search tree;

[0231] Determine data of an input item in a cost estimation function according to the shortest distance, the current candidate grid, the previous path node, the current starting point position and the destination position, and substitute the data into the input item to perform cost calculation;

[0232] The current path node is determined from the current candidate grid according to the calculated cost.

[0233] In some embodiments, the planning unit 204 is specifically configured to:

[0234] Determine the diameter of the safety circle required for the mobile vehicle during driving;

[0235] Determine a target circle with the position of the current candidate grid as the center and the diameter of the safety circle as the diameter;

[0236] According to the center and radius of each node in the search tree, a leaf node intersecting with the target circle is determined from the search tree;

[0237] The minimum Euclidean distance between the current candidate grid and the obstacle grid in the intersecting leaf node is calculated to obtain the shortest distance between the current candidate grid and the nearest obstacle.

[0238] It should be noted that the specific details of each module unit in the above-mentioned driving path planning device 200 have been described in detail in the embodiment of the above-mentioned driving path planning method, and will not be repeated here.

[0239] In some embodiments, the driving path planning device in the embodiments of the present application may be an electronic device, or a component in an electronic device, such as an integrated circuit or a chip. The electronic device may be an autonomous driving vehicle.

[0240] In some embodiments, Figure 8 As shown, an embodiment of the present application also provides an electronic device 300, including a processor 301, a memory 302, and a computer program stored in the memory 302 and executable on the processor 301. When the program is executed by the processor 301, each process of the above-mentioned driving path planning method embodiment is implemented, and the same technical effect can be achieved. To avoid repetition, it will not be described here.

[0241] It should be noted that the electronic devices in the embodiments of the present application include the mobile electronic devices and non-mobile electronic devices mentioned above.

[0242] Fig. 9 A schematic diagram of the hardware structure of an electronic device to implement an embodiment of the present application.

[0243] The electronic device 400 includes but is not limited to: a radio frequency unit 401, a network module 402, an audio output unit 403, an input unit 404, a sensor 405, a display unit 406, a user input unit 407, an interface unit 408, a memory 409, a processor 410 and other components.

[0244] Those skilled in the art will appreciate that the electronic device 400 may also include a power source (such as a battery) for supplying power to various components, and the power source may be logically connected to the processor 410 through a power management system, thereby implementing functions such as managing charging, discharging, and power consumption management through the power management system. Fig. 9 The electronic device structure shown in the figure does not constitute a limitation on the electronic device. The electronic device may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently, which will not be described in detail here.

[0245] It should be understood that in the embodiment of the present application, the input unit 404 may include a graphics processor (GPU) 4041 and a microphone 4042, and the graphics processor 4041 processes the image data of the static picture or video obtained by the image capture device (such as a camera) in the video capture mode or the image capture mode. The display unit 406 may include a display panel 4061, and the display panel 4061 may be configured in the form of a liquid crystal display, an organic light emitting diode, etc. The user input unit 407 includes a touch panel 4071 and at least one of other input devices 4072. The touch panel 4071 is also called a touch screen. The touch panel 4071 may include two parts: a touch detection device and a touch controller. Other input devices 4072 may include, but are not limited to, a physical keyboard, function keys (such as a volume control key, a switch key, etc.), a trackball, a mouse, and a joystick, which will not be repeated here.

[0246] The memory 409 can be used to store software programs and various data. The memory 409 may mainly include a first storage area for storing programs or instructions and a second storage area for storing data, wherein the first storage area may store an operating system, an application program or instructions required for at least one function (such as a sound playback function, an image playback function, etc.), etc. In addition, the memory 409 may include a volatile memory or a non-volatile memory, or the memory 409 may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), a static random access memory (SRAM), a dynamic random access memory (DRAM), a synchronous dynamic random access memory (SDRAM), a double data rate synchronous dynamic random access memory (DDRSDRAM), an enhanced synchronous dynamic random access memory (ESDRAM), a synchronous link dynamic random access memory (SLDRAM) and a direct memory bus random access memory (DRRAM). The memory 409 in the embodiment of the present application includes but is not limited to these and any other suitable types of memory.

[0247] The processor 410 may include one or more processing units; the processor 410 integrates an application processor and a modem processor, wherein the application processor mainly processes operations related to an operating system, a user interface, and application programs, and the modem processor mainly processes wireless communication signals, such as a baseband processor. It is understandable that the modem processor may not be integrated into the processor 410.

[0248] An embodiment of the present application also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the various processes of the above-mentioned driving path planning method embodiment are implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0249] The processor is the processor in the electronic device described in the above embodiment. The readable storage medium includes a computer readable storage medium, such as a computer read-only memory ROM, a random access memory RAM, a magnetic disk or an optical disk.

[0250] An embodiment of the present application also provides a computer program product, including a computer program, which implements the above-mentioned driving path planning method when executed by a processor.

[0251] The processor is the processor in the electronic device described in the above embodiment. The readable storage medium includes a computer readable storage medium, such as a computer read-only memory ROM, a random access memory RAM, a magnetic disk or an optical disk.

[0252] It should be noted that, in this article, the terms "comprise", "include" or any other variant thereof are intended to cover non-exclusive inclusion, so that the process, method, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise one..." do not exclude the presence of other identical elements in the process, method, article or device including the element. In addition, it should be pointed out that the scope of the methods and devices in the embodiments of the present application is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in reverse order according to the functions involved, for example, the described method may be performed in an order different from that described, and various steps may also be added, omitted, or combined. In addition, the features described with reference to certain examples may be combined in other examples.

[0253] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a computer software product, which is stored in a storage medium (such as ROM / RAM, a magnetic disk, or an optical disk), and includes a number of instructions for enabling a terminal (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present application.

[0254] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present application, ordinary technicians in this field can also make many forms without departing from the purpose of the present application and the scope of protection of the claims, all of which are within the protection of the present application.

[0255] The terms "first", "second", etc. in the specification and claims of the present application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable under appropriate circumstances, so that the embodiments of the present application can be implemented in an order other than those illustrated or described here, and the objects distinguished by "first", "second", etc. are generally of one type, and the number of objects is not limited. For example, the first object can be one or more. In addition, "and / or" in the specification and claims represents at least one of the connected objects, and the character " / " generally indicates that the objects associated with each other are in an "or" relationship.

[0256] In the description of the present application, “plurality” means two or more.

[0257] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "illustrative embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials, or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0258] Although the embodiments of the present application have been shown and described, those skilled in the art will appreciate that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present application, and that the scope of the present application is defined by the claims and their equivalents.

Claims

1. A driving path planning method, characterized in that: include: Obtaining a grid map of a driving environment and an obstacle grid on the grid map; Dividing all the obstacle grids into a plurality of obstacle grid groups; Creating a retrieval tree corresponding to each obstacle grid group, wherein each tree node in the retrieval tree corresponds to a circular area; Determining a driving path of the mobile vehicle based on the starting position, the destination position and the search tree of the mobile vehicle; The step of dividing all the obstacle grids into a plurality of obstacle grid groups includes: Performing principal component analysis on all the obstacle grids to determine the principal component direction, and taking the direction perpendicular to the principal component direction as the extension direction of the segmentation line to be determined; Determining a segmentation line according to the principal component direction, the extension direction and the obstacle grid; All the obstacle grids are divided into a plurality of obstacle grid groups according to the dividing lines.

2. The driving path planning method according to claim 1, characterized in that: The determining of the segmentation line according to the principal component direction, the extension direction and the obstacle grid comprises: Determining an intersection point between the to-be-determined segmentation line and the principal component direction according to the obstacle grid; A straight line passing through the intersection point and extending along the extending direction is determined as a dividing line.

3. The driving path planning method according to claim 1, characterized in that: The step of creating a search tree corresponding to each obstacle grid group includes: Determine a minimum circle that includes all the obstacle grids in the obstacle grid group; From the obstacle grid group, select the obstacle grid farthest from the center of the minimum circle as a first reference grid, and select the obstacle grid farthest from the first reference grid as a second reference grid; A corresponding retrieval tree is created according to the first reference grid, the second reference grid and the obstacle grid group.

4. The driving path planning method according to claim 3, characterized in that: The step of creating a corresponding search tree according to the first reference grid, the second reference grid and the obstacle grid group includes: Determine a first distance between each obstacle grid in the obstacle grid group and the first reference grid, and a second distance between each obstacle grid and the second reference grid; Dividing the obstacle grid group into two subgroups according to the first distance and the second distance; A corresponding search tree is created according to the subgroup, the smallest circle including all the obstacle grids in the subgroup, the obstacle grid group, and the smallest circle including all the obstacle grids in the obstacle grid group.

5. The driving path planning method according to claim 1, characterized in that: The step of determining the driving path of the mobile vehicle based on the starting position, the destination position and the search tree of the mobile vehicle comprises: According to the starting position, destination position and the retrieval tree of the mobile vehicle, a driving path of the mobile vehicle is planned on the grid map through a cost estimation function, wherein an input item of the cost estimation function includes the shortest distance between a path node and a nearest obstacle.

6. The driving path planning method according to claim 5, characterized in that: The method of planning a driving path of the mobile vehicle on the grid map by using a cost estimation function according to the starting position, the destination position and the search tree of the mobile vehicle comprises: Determining a current planned driving path, the initially planned driving path being a starting position of the moving vehicle; Taking the path node at the end of the planned driving path as the current starting point of the mobile vehicle; Determining a corresponding first driving path on the grid map by using an RS curve method according to the current starting point and the destination position; When the first driving path does not collide with an obstacle, the planned driving path and the first driving path are spliced ​​to obtain a final driving path of the mobile vehicle; When the first driving path collides with the obstacle, a second driving path of the mobile vehicle is planned on the grid map through a cost estimation function according to the current starting point, the destination position and the retrieval tree, and a final driving path of the mobile vehicle is determined according to the planned driving path and the second driving path.

7. The driving path planning method according to claim 6, characterized in that: The step of planning a second driving path of the mobile vehicle on the grid map by using a cost estimation function according to the current starting point, the destination position and the search tree comprises: When planning a current path node in the second driving path on the grid map, determining a plurality of current candidate grids; Determine the shortest distance between the current candidate grid and the nearest obstacle according to the search tree; Determine data of input items in a cost estimation function according to the shortest distance, the current candidate grid, the previous path node, the current starting point and the destination position, and substitute the data into the input items to perform cost calculation; A current path node is determined from the current candidate grid according to the calculated cost.

8. The driving path planning method according to claim 7, characterized in that: The determining, according to the search tree, the shortest distance between the current candidate grid and the nearest obstacle comprises: Determining the required safety circle diameter of the mobile vehicle during driving; Determine a target circle with the position of the current candidate grid as the center and the diameter of the safety circle as the diameter; Determine, from the search tree, a leaf node intersecting with the target circle according to the center and radius corresponding to each node in the search tree; The minimum Euclidean distance between the current candidate grid and the obstacle grid in the intersecting leaf node is calculated to obtain the shortest distance between the current candidate grid and the nearest obstacle.

9. A driving path planning device, characterized in that: include: An acquisition unit, used to acquire a grid map of the driving environment and obstacle grids on the grid map; A division unit, used for dividing all the obstacle grids into a plurality of obstacle grid groups, wherein the division of all the obstacle grids into a plurality of obstacle grid groups comprises: performing principal component analysis on all the obstacle grids to determine a principal component direction, and taking a direction perpendicular to the principal component direction as an extension direction of a segmentation line to be determined; determining a segmentation line according to the principal component direction, the extension direction and the obstacle grids; and dividing all the obstacle grids into a plurality of obstacle grid groups according to the segmentation line; A creating unit, used for creating a retrieval tree corresponding to each obstacle grid group, wherein each tree node in the retrieval tree corresponds to a circular area; A planning unit is used to determine a driving path of the mobile vehicle based on a starting position, a destination position and the retrieval tree of the mobile vehicle.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the driving path planning method according to any one of claims 1 to 8 is implemented.

11. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the driving path planning method according to any one of claims 1 to 8 is implemented.

12. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the driving path planning method according to any one of claims 1 to 8 is implemented.