Path planning method and device, computer program product and electronic equipment

Abstract

The invention relates to the technical field of computers, in particular to a path planning method and device, a computer program product and electronic equipment. The path planning method comprises the following steps: acquiring a path prediction area comprising a starting node and a destination node, wherein the path prediction area further comprises a plurality of prediction nodes; a path planning algorithm model is constructed, the path planning algorithm model is obtained by improving an open list of the heuristic search algorithm model based on a binary tree structure, and a root node of the binary tree structure is set as a prediction node corresponding to the lowest path prediction cost in the open list; on the basis of the path prediction cost of the prediction nodes in the path prediction area, path planning is carried out in the forward direction and the reverse direction through a path planning algorithm model, and a reference planning path is determined according to forward and reverse planning results; and performing path smoothing processing on the reference planning path to obtain a target planning path. The path planning efficiency and the continuity of the path planning result can be improved.

Description

Technical Field

[0001] The present disclosure relates to the field of computer technology, and more specifically, to a path planning method, a path planning device, a computer program product, and an electronic device. Background Art

[0002] In recent years, autonomous driving has been widely used in logistics, rescue, exploration, military and other fields. Autonomous driving vehicles can move to the destination according to the planned path. Whether the planned path is accurate affects the performance of autonomous driving. The current path planning method has the problems of slow search speed and many search nodes, which affects the efficiency of path planning to a certain extent. In addition, the path planning results have a "sawtooth" path, which leads to unstable movement and overdriving of autonomous driving vehicles.

[0003] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present disclosure, and therefore may include information that does not constitute the prior art known to ordinary technicians in the field. Summary of the invention

[0004] The purpose of the present disclosure is to provide a path planning method and device, a computer program product and an electronic device, thereby improving the path planning efficiency and the continuity of the path planning results at least to a certain extent.

[0005] Other features and advantages of the present disclosure will become apparent from the following detailed description, or may be learned in part by the practice of the present disclosure.

[0006] According to one aspect of the present disclosure, a path planning method is provided, including: obtaining a path prediction area including a start node and a destination node, wherein the path prediction area also includes a plurality of prediction nodes; constructing a path planning algorithm model, wherein the path planning algorithm model is obtained by improving an open list of a heuristic search algorithm model based on a binary tree structure, wherein the root node of the binary tree structure is set to a prediction node corresponding to a lowest path prediction cost in the open list; based on the path prediction costs of the prediction nodes in the path prediction area, performing path planning from forward and reverse directions respectively using the path planning algorithm model, so as to determine a reference planning path according to the forward and reverse planning results; and performing path smoothing processing on the reference planning path to obtain a target planning path.

[0007] In an exemplary embodiment of the present disclosure, the heuristic search algorithm model includes an open list for storing unpredicted nodes in a path planning process and a closed list for storing predicted nodes in a path planning process; constructing a path planning algorithm model, including: determining the path prediction cost of the unpredicted nodes; constructing a binary tree structure based on the open list, so that the root node of the binary tree structure is always the predicted node corresponding to the lowest path prediction cost in the open list, and the path prediction cost of the child node is greater than or equal to the path prediction cost of the corresponding parent node.

[0008] In an exemplary embodiment of the present disclosure, based on the path prediction cost of the prediction node in the path prediction area, a path planning algorithm model is used to perform path planning from forward and reverse directions respectively, so as to determine a reference planning path according to the forward and reverse planning results, including: based on the path prediction cost of the prediction node in the path prediction area, forward path planning is performed from the starting node to the destination node using the path planning algorithm model, and reverse path planning is performed from the destination node to the starting node using the path planning algorithm model; wherein the path planning algorithm model includes a first binary tree structure corresponding to the forward path planning, a second binary tree structure corresponding to the reverse path planning, and a closed list corresponding to the forward path planning and the reverse path planning.

[0009] In an exemplary embodiment of the present disclosure, based on the path prediction cost of the prediction node in the path prediction area, the path planning algorithm model is used to perform path planning from forward and reverse directions respectively, so as to determine the reference planning path according to the forward and reverse planning results, including: in the forward path planning process, determining the first prediction node corresponding to the lowest path prediction cost based on the first binary tree structure; in the reverse path planning process, determining the second prediction node corresponding to the lowest path prediction cost based on the second binary tree structure; if the first prediction node is added to the closed list in the reverse path planning process, then the planned path from the first prediction node to the starting node in the forward path planning process is spliced ​​with the planned path from the destination node to the first prediction node in the reverse path planning process to obtain the reference planning path; or, if the second prediction node is added to the closed list in the forward path planning process, then the planned path from the destination node to the second prediction node in the reverse path planning process is spliced ​​with the planned path from the starting node to the second prediction node in the forward path planning process to obtain the reference planning path.

[0010] In an exemplary embodiment of the present disclosure, the method further includes: if the forward and reverse planning results obtained through the forward path planning process and the reverse path planning process respectively do not contain overlapping prediction nodes, then the forward planning result or the reverse planning result is determined as a reference planning path.

[0011] In an exemplary embodiment of the present disclosure, a forward planning result or a reverse planning result is determined as a reference planning path, including: comparing the total path prediction cost corresponding to the forward planning result with the total path prediction cost corresponding to the reverse planning result; and determining the forward planning result or the reverse planning result as the reference planning path based on the comparison result.

[0012] In an exemplary embodiment of the present disclosure, a reference planned path is smoothed to obtain a target planned path, including: if the number of nodes in the reference planned path is greater than a preset number threshold, a first smoothing strategy is used to eliminate redundant nodes in the reference planned path, and a second smoothing strategy is used to smooth the reference planned path after eliminating the redundant nodes; if the number of nodes in the reference planned path is less than or equal to the preset number threshold, a second smoothing strategy is used to smooth the reference planned path; wherein the second smoothing strategy is used to eliminate nodes based on the distance between adjacent nodes.

[0013] In an exemplary embodiment of the present disclosure, a first smoothing strategy is adopted to eliminate redundant nodes in a reference planning path, including: looping the following process in the order from the starting node to the destination node until the number of nodes in the reference planning path meets the preset number of nodes: obtaining a ternary continuous node group of the current node to be checked from the reference planning path; if the ternary continuous node group meets the node elimination strategy, the second node in the ternary continuous node group is eliminated, the reference planning path is updated according to the elimination result, and the step of obtaining the ternary continuous node group of the current node to be checked from the reference planning path is returned; if the ternary continuous node group does not meet the node elimination strategy, the current node to be checked is updated based on the reference planning path, and the step of obtaining the ternary continuous node group of the current node to be checked from the reference planning path is returned.

[0014] In an exemplary embodiment of the present disclosure, a node removal strategy includes: determining that a third node in a ternary continuous node group is directly visible relative to a first node based on position information of each node in the ternary continuous node group; and a distance between the first node and the third node is less than or equal to the sum of a distance between the first node and the second node and a distance between the second node and the third node.

[0015] In an exemplary embodiment of the present disclosure, for the reference planned path after smoothing by the second smoothing strategy, it also includes: interpolating adjacent nodes in the reference planned path after smoothing by the second smoothing strategy to obtain continuous path points; connecting the obtained continuous path points to obtain the target planned path.

[0016] According to one aspect of the present disclosure, a path planning device is provided, including: an area acquisition module, used to acquire a path prediction area including a start node and a destination node, wherein the path prediction area also includes multiple prediction nodes; an algorithm construction module, used to construct a path planning algorithm model, wherein the path planning algorithm model is obtained by improving an open list of a heuristic search algorithm model based on a binary tree structure, wherein the root node of the binary tree structure is set to a prediction node corresponding to a lowest path prediction cost in the open list; a path planning module, used to perform path planning from a forward and a reverse direction respectively based on the path prediction costs of the prediction nodes in the path prediction area using the path planning algorithm model, so as to determine a reference planning path according to the forward and reverse planning results; and a path smoothing module, used to perform path smoothing processing on the reference planning path to obtain a target planning path.

[0017] According to one aspect of the present disclosure, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, any one of the above methods is implemented.

[0018] According to one aspect of the present disclosure, an electronic device is provided, comprising: a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to perform any one of the above methods by executing the executable instructions.

[0019] The path planning method in the exemplary embodiment of the present disclosure, on the one hand, obtains the path prediction area including the starting node and the destination node, and constructs the path planning algorithm model. The path planning algorithm model is obtained by improving the open list of the heuristic search algorithm model based on the binary tree structure, so that the root node of the binary tree structure is set to the prediction node corresponding to the lowest path prediction cost in the open list. The path planning algorithm model based on the improved heuristic search algorithm model only needs to target the root node of the binary tree when obtaining and deleting nodes during path planning. Based on the binary tree structure, the time complexity required for node comparison can be reduced, and the path planning efficiency is improved. On the other hand, the path planning algorithm model is used to perform path planning from the forward and reverse directions respectively, so as to determine the reference planning path according to the forward and reverse planning results. Based on the bidirectional search strategy, the optimal path can be found faster, and the planning efficiency of the path can be further improved. On the other hand, the target planning path obtained by path smoothing the reference planning path can eliminate the "sawtooth" path existing in the planning result based on the path prediction cost during the path planning process, avoid the situation of path discontinuity, and thus avoid the problem of motion instability and overdriving caused by the path non-smoothness.

[0020] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The above and other objects, features and advantages of the exemplary embodiments of the present disclosure will become readily understood by reading the following detailed description with reference to the accompanying drawings, in which several embodiments of the present disclosure are shown in an exemplary and non-limiting manner.

[0022] Figure 1 An application environment according to an exemplary embodiment of the present disclosure is shown.

[0023] Figure 2 A flow chart of a path planning method according to an exemplary embodiment of the present disclosure is shown.

[0024] Figure 3 A schematic diagram of a grid map according to an exemplary embodiment of the present disclosure is shown.

[0025] Figure 4 A flow chart of a path planning method based on binary tree improvement according to an exemplary embodiment of the present disclosure is shown.

[0026] Figure 5 A flowchart of determining a reference planning path by performing path planning in forward and reverse directions respectively using a path planning algorithm model according to an exemplary embodiment of the present disclosure is shown.

[0027] Figure 6 A schematic diagram of path smoothing according to an exemplary embodiment of the present disclosure is shown.

[0028] Figure 7 A flow chart of performing path smoothing processing on a reference planned path according to an exemplary embodiment of the present disclosure is shown.

[0029] Figure 8 A schematic diagram of the composition of a path planning device according to an exemplary embodiment of the present disclosure is shown.

[0030] Fig. 9 A block diagram of an electronic device according to an exemplary embodiment of the present disclosure is shown.

[0031] In the drawings, the same or corresponding reference numerals represent the same or corresponding parts. DETAILED DESCRIPTION

[0032] The exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, the exemplary embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that the present disclosure will be more comprehensive and complete and the concepts of the exemplary embodiments will be fully conveyed to those skilled in the art. The same reference numerals in the figures represent the same or similar structures, and thus their detailed description will be omitted.

[0033] In addition, the described features, structures or characteristics may be combined in one or more embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the embodiments of the present disclosure. However, those skilled in the art will appreciate that the technical solutions of the present disclosure may be practiced without one or more of the specific details, or other methods, components, devices, steps, etc. may be adopted. In other cases, known structures, methods, devices, implementations or operations are not shown or described in detail to avoid blurring the various aspects of the present disclosure.

[0034] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities may be implemented in software form, or these functional entities or parts of functional entities may be implemented in one or more software hardened modules, or these functional entities may be implemented in different networks and / or processor devices and / or microcontroller devices.

[0035] There are many path planning methods. For static environments, the A* algorithm is one of the most effective methods for searching paths, so it is widely used in path planning simulation experiments. As a heuristic search, the A* algorithm returns an optimal path with the minimum cost as long as there is a path, thereby ensuring integrity. However, the traditional A* algorithm is limited to generating piecewise linear paths, which are neither smooth nor continuous. Although a three-layer hierarchical architecture path planning algorithm has been developed in the related art, in which obstacles are confined to a box. When the A* algorithm is applied locally in the second layer, the starting point of the search path is the current position of the autonomous driving device, and the end point is one of the vertices of the obstacle bounding box. There are still problems such as slow search speed, many search nodes, and long search time. In addition, the resulting path planning is not smooth, which may cause unstable motion and overdriving of the unmanned driving device. In view of this, an exemplary embodiment of the present disclosure provides a path planning method, which improves the heuristic search algorithm model based on a binary tree structure, and performs a bidirectional search strategy from forward and reverse directions based on the obtained path planning algorithm model, thereby improving the path planning efficiency. At the same time, the obtained reference planning path is subjected to path smoothing processing to make the target planning path more continuous and smooth, thereby enabling the unmanned driving equipment that performs path planning based on the path planning method of the exemplary embodiment of the present disclosure to operate stably and efficiently.

[0036] It should be noted that the path planning method and device provided by the exemplary embodiments of the present disclosure can be used in the fields of computer technology and autonomous driving technology, for example, in the fields of positioning and navigation. The path planning method and device provided by the exemplary embodiments of the present disclosure can also be used in any field other than the fields of computer technology and autonomous driving technology, for example, in the field of warehouse management, by performing path planning on a mobile robot to achieve the acquisition / installation of goods. The application field of the path planning method and device provided by the exemplary embodiments of the present disclosure is not limited.

[0037] The path planning method provided by the exemplary embodiment of the present disclosure can be applied to Figure 1 In the application environment shown, the terminal 101 communicates with the server 102 via a network. The data storage system can store data that the server 102 needs to process. The data storage system can be integrated on the server 102, or placed on the cloud or other network servers.

[0038] In an exemplary embodiment, the path planning method provided by the exemplary embodiment of the present disclosure may be executed by the server 102, and the corresponding path planning device is disposed in the server 102. Correspondingly, in this mode of execution by the server 102, the server 102 may start to execute the steps in the technical solution of the exemplary embodiment of the present disclosure in response to a trigger command, wherein the trigger command may be sent by a terminal used by a user, or may be triggered locally by the server in response to some automation events.

[0039] The server 102 may be an independent physical server, a server cluster or a distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. The server 102 may perform background tasks.

[0040] Furthermore, in another exemplary embodiment, the terminal 101 may also have similar functions to the server 102 , so as to execute the path planning method provided by the exemplary embodiments of the present disclosure.

[0041] The terminal 101 may be an autonomous driving device, a smart phone, a tablet computer, a laptop computer, a desktop computer, a smart speaker, etc. The terminal 101 may also be referred to as a mobile terminal, a terminal device, a mobile device, etc. The exemplary embodiments of the present disclosure do not limit the type of the terminal 101.

[0042] In addition, the technical solution of the exemplary embodiment of the present disclosure can also be executed collaboratively by the terminal 101 and the server 102. In this way of collaborative execution by the terminal 101 and the server 102, some steps in the technical solution provided by the exemplary embodiment of the present disclosure are executed by the terminal 101, while other steps are executed by the server 102. For example, the server 102 obtains a path prediction area including a start node and a destination node, and constructs a path planning algorithm model, and sends the path prediction area and the path planning algorithm model to the terminal 101, and the terminal 101 performs path planning from the forward and reverse directions respectively based on the path prediction cost of the predicted node in the path prediction area using the path planning algorithm model, so as to determine the reference planning path according to the forward and reverse planning results, and perform path smoothing processing on the reference planning path to obtain the target planning path.

[0043] It should be noted that in this mode of collaborative execution by the terminal 101 and the server 102, the steps respectively executed by the terminal 101 and the server 102 can be dynamically adjusted according to actual conditions, and there is no special restriction on this.

[0044] The terminal 101 and the server 102 may be connected directly or indirectly via wireless communication, and the exemplary embodiments of the present disclosure do not impose any special limitation thereto.

[0045] It should be noted that the sequence numbers of the operations in the following method are only used as representations of the operations for the purpose of description, and should not be regarded as representing the execution order of the operations. Unless explicitly stated, the method does not need to be executed completely in the order shown.

[0046] Figure 2 FIG. 1 is a flow chart of a path planning method according to an exemplary embodiment of the present disclosure. Figure 2 As shown, the path planning method includes steps S210 to S240, which are specifically as follows:

[0047] In step S210, a path prediction area including a start node and a destination node is obtained, and the path prediction area also includes a plurality of prediction nodes.

[0048] In the exemplary embodiments of the present disclosure, the path planning is global path planning, which specifically refers to facing the destination and requiring the autonomous driving device (such as an unmanned vehicle) to find a complete path connecting the starting point and the end point according to the known environment. Among them, the environment in which the autonomous driving device is located can be virtualized by map, that is, a grid map is generated according to the location of the autonomous driving device and the environment in which it is located to obtain a path prediction area. For example, the environment is divided into a plurality of small squares through a grid map, so that the map can be represented by a two-dimensional array. Of course, the exemplary embodiments of the present disclosure do not limit the way of dividing the grid map, for example, it can also be divided into hexagonal grids, rectangular grids, etc., which can be flexibly set according to needs. The multiple prediction nodes in the path prediction area refer to the nodes in the grid map except the starting node and the destination node. Based on these nodes, an optimal path from the starting node to the destination node can be obtained in the path planning process. Of course, the number and distribution of the prediction nodes in the path prediction area are also related to the way of generating the grid map, and there is no limitation on this.

[0049] For example, Figure 3 A schematic diagram of a grid map is shown, Figure 3 As shown, the "a" area represents the starting node, the "b" area represents the destination node, the "c" area represents the obstacles, the "d" area represents the predicted nodes in the open list to be searched (unpredicted nodes), the "e" area represents the predicted nodes in the closed list that have been searched (predicted nodes), and the "x" area represents the currently searched path.

[0050] In step S220, a path planning algorithm model is constructed. The path planning algorithm model is obtained by improving the open list of the heuristic search algorithm model based on a binary tree structure. The root node of the binary tree structure is set to the prediction node corresponding to the lowest path prediction cost in the open list.

[0051] In the exemplary embodiment of the present disclosure, the path planning algorithm model is a model for path planning, which is obtained by improving the heuristic search algorithm model. The heuristic search algorithm is to evaluate each search position (node) in the state space, obtain the best position, and then search from this position until the target position, including A* algorithm, genetic algorithm, ant colony algorithm and simulated annealing algorithm, etc. The heuristic search algorithm model of the exemplary embodiment of the present disclosure is explained by taking the A* search algorithm as an example.

[0052] The A* algorithm introduces global information when checking each possible node in the shortest path, estimates the distance between the current node and the end point, and uses it as a measure to evaluate the possibility that the node is on the shortest route. The A* algorithm needs to define two empty lists, called open-list and closed-list. The open-list stores nodes that have been preprocessed but not investigated, and the closed-list stores nodes that have been investigated. The most time-consuming part of the A* algorithm is to traverse the nodes in the open list to obtain the node with the lowest path prediction cost. As the search proceeds, the number of nodes in the open list gradually increases, resulting in a significant increase in search time. Among them, the path prediction cost refers to the total path consumption from the node to the destination node. For example, G represents the path cost of moving from the starting node to the specified square (node) in the path prediction area (such as moving up, down, left, right, or along the diagonal direction), and H represents the path cost of moving from the specified square (node) to the destination node. Therefore, the path prediction cost of the specified square (node) is F=G+H.

[0053] The binary tree structure refers to an ordered tree in which the degree of the nodes in the tree is not greater than 2. The open list of the A* algorithm is improved through the binary tree structure, so that the root node of the binary tree structure is set to the prediction node corresponding to the lowest path prediction cost in the open list. Each time the prediction node corresponding to the lowest path prediction cost is quickly extracted from the open list, the lowest path prediction cost can be directly determined based on the root node of the binary tree, reducing the time complexity of the search.

[0054] In step S230, based on the path prediction costs of the prediction nodes in the path prediction area, the path planning algorithm model is used to perform path planning from the forward and reverse directions respectively, so as to determine a reference planning path according to the forward and reverse planning results.

[0055] In an exemplary embodiment of the present disclosure, a bidirectional search strategy is adopted based on an improved path planning algorithm model, that is, it is necessary to traverse from the starting node to the destination node, and also from the destination node to the starting node, so that the reference planning path can be determined according to the obtained positive planning results and negative planning results.

[0056] In some scenarios, by performing path planning from both the forward and reverse directions, the optimal path to a node can be searched for simultaneously in the forward and negative path planning processes, thereby determining the reference planning path without having to traverse all predicted nodes in the forward or negative direction, further reducing the number of search nodes and improving path planning efficiency. Of course, in some scenarios, there may also be bidirectional path planning that finds its own reference planning path, that is, the forward and reverse path planning do not traverse to the same node, in which case the final reference planning path can be determined based on the forward and reverse path planning results.

[0057] It should be understood that based on the improved path planning algorithm, in the process of forward and reverse path planning, each time the predicted node with the lowest path prediction cost is obtained, it can be obtained based on the binary tree structure corresponding to the open list, thereby reducing the time complexity.

[0058] In step S240, path smoothing is performed on the reference planned path to obtain a target planned path.

[0059] In an exemplary embodiment of the present disclosure, the trajectory of the reference planning path is composed of many short line segments connecting a series of points, the path is discontinuous in navigation, there is a "zigzag" path, and the turning cost is sacrificed. Therefore, by performing path smoothing on the reference planning path, the target planning path is made smooth and continuous, thereby improving motion stability and avoiding problems such as overdriving at the path position of the sharp turning point.

[0060] The path planning method in the exemplary embodiment of the present disclosure, on the one hand, obtains the path prediction area including the starting node and the destination node, and constructs the path planning algorithm model. The path planning algorithm model is obtained by improving the open list of the heuristic search algorithm model based on the binary tree structure, so that the root node of the binary tree structure is set to the prediction node corresponding to the lowest path prediction cost in the open list. The path planning algorithm model based on the improved heuristic search algorithm model only needs to target the root node of the binary tree when obtaining and deleting nodes during path planning. Based on the binary tree structure, the time complexity required for node comparison can be reduced, and the path planning efficiency is improved. On the other hand, the path planning algorithm model is used to perform path planning from the forward and reverse directions respectively, so as to determine the reference planning path according to the forward and reverse planning results. Based on the bidirectional search strategy, the optimal path can be found faster, and the planning efficiency of the path can be further improved. On the other hand, the target planning path obtained by path smoothing the reference planning path can eliminate the "sawtooth" path existing in the planning result based on the path prediction cost during the path planning process, avoid the situation of path discontinuity, and thus avoid the problem of motion instability and overdriving caused by the path non-smoothness.

[0061] In an exemplary embodiment, an implementation method for constructing a path algorithm model is provided. As can be seen above, the A* algorithm includes an open list for storing unpredicted nodes in the path planning process and a closed list for storing predicted nodes in the path planning process. The key to constructing the path planning algorithm model is to use a binary tree structure to improve the structure of the open list. Specifically, constructing the path planning algorithm model may include:

[0062] Determine the path prediction cost of the unpredicted nodes;

[0063] A binary tree structure is constructed according to the open list, so that the root node of the binary tree structure is always the prediction node corresponding to the lowest path prediction cost in the open list, and the path prediction cost of the child node is greater than or equal to the path prediction cost of the corresponding parent node.

[0064] Among them, considering that the node with the lowest cost, that is, the predicted node corresponding to the lowest path prediction cost, needs to be quickly extracted from the open list each time during the path planning process, and whether other nodes are sorted is not important, the exemplary embodiment of the present disclosure uses a binary tree structure to improve the structure of the open list. The root node of the binary tree structure is always the predicted node corresponding to the lowest path prediction cost in the open list. It can be understood that in the path planning process, as the type and number of nodes in the open list change, the binary tree structure is updated according to the open list, so that the root node of the binary tree structure is always the predicted node corresponding to the lowest path prediction cost in the open list.

[0065] As an example, the identification information of the prediction node is used as the key, and the path prediction cost of the prediction node is used as the key value to determine the two-dimensional data, and the binary tree is realized through the two-dimensional data, where the item with the smallest key value is always the root node of the binary tree structure, and the key values ​​of the two child nodes of each node are greater than or equal to the parent node. For example, if the root node is stored in array [i], its child nodes are stored in array [2i] and array [2i+1] respectively.

[0066] By improving the structure of the open list, for an array consisting of n elements, if the nodes are still traversed in the existing way, the time complexity of traversing the nodes is O(n), while the time complexity of using the open list updated based on the binary tree structure is O(1). Accordingly, each time a node in the open list is deleted, only the predicted node corresponding to the root node needs to be deleted, which can reduce processing time and improve path planning efficiency.

[0067] In an exemplary embodiment, a path planning method based on a bidirectional search strategy is provided. Based on the path prediction cost of the prediction node in the path prediction area, the path planning algorithm model is used to perform path planning from forward and reverse directions respectively, so as to determine a reference planning path according to the forward and reverse planning results, including:

[0068] Based on the path prediction cost of the prediction node in the path prediction area, the forward path planning is performed from the starting node to the destination node using the path planning algorithm model, and the reverse path planning is performed from the destination node to the starting node using the path planning algorithm model;

[0069] The path planning algorithm model includes a first binary tree structure corresponding to the forward path planning, a second binary tree structure corresponding to the reverse path planning, and a closed list table corresponding to the forward path planning and the reverse path planning.

[0070] Specifically, when forward and reverse path planning are performed simultaneously, the path planning algorithm model includes a first binary tree structure corresponding to the forward path planning, which is used to store the predicted nodes in the forward path planning process, and the root node of the first binary tree structure always corresponds to the predicted node corresponding to the lowest path prediction cost in the open list of the forward path planning process. Correspondingly, the second binary tree structure corresponding to the reverse path planning is used to store the predicted nodes in the reverse path planning process, and the root node of the second binary tree structure always corresponds to the predicted node corresponding to the lowest path prediction cost in the open list of the reverse path planning process, and the forward path planning process and the reverse path planning process share a closed list.

[0071] Among them, the forward path planning process and the reverse path planning process share a closed list. When a predicted node in the path prediction area is judged based on the closed list during the bidirectional search process, both the forward and reverse path planning processes can obtain the optimal path, thereby terminating the node traversal in advance and improving the path planning efficiency.

[0072] As an example, Figure 4 A flowchart of a path planning method based on a binary tree improvement is shown below. Figure 4 , taking the forward path planning process as an example for explanation. It should be understood that the core idea of ​​the reverse path planning process is consistent with that of the forward path planning process.

[0073] First, add the starting node S0 to the open list and calculate f(S0).

[0074] The binary tree structure is updated according to the open list so that the root node of the binary tree structure is the prediction node corresponding to the lowest path prediction cost. The method for calculating f(S0) can refer to the method for calculating the path prediction cost (F=G+H) of the existing A* algorithm, and there is no limitation on this.

[0075] Next, determine whether the open list is empty.

[0076] If the open list is empty, the current path planning fails. If the open list is not empty, the node corresponding to the lowest path prediction cost is quickly determined based on the binary tree structure corresponding to the open list, recorded as n*.

[0077] Next, determine whether n* is the destination node Sg. If n* is the destination node Sg, obtain the reference planning path, i.e., S0 to n*(Sg). If n* is not the destination node Sg, expand n* to generate all the child nodes of n* and obtain the child node set.

[0078] Among them, the predicted nodes (such as adjacent nodes) around n* can be obtained to generate a child node set. Of course, the method of expanding n* can also be selected according to the actual scene requirements, and there is no specific restriction on this.

[0079] Then, calculate the path prediction cost f(pre) for the prediction nodes in the child node set of n*. Here, pre represents the current prediction node in the child node set.

[0080] Furthermore, determine whether the current prediction node pre is in the open list and the closed list. Among them, if pre is not in the open list and the closed list, no operation of updating the value is performed, and set the pointer to make pre point to n* (that is, update the parent node of pre to n*). If pre is in the open list or the closed list, determine whether g(new_pre) is less than g(ole_pre). Here, g(new_pre) is the path cost of S0→n*→pre (that is, the path passing through n*), and g(ole_pre) is the path cost of S0→pre. Among them, if g(new_pre) < g(ole_pre), it means that the new path has a lower path cost, then replace g(ole_pre) with g(new_pre), and set the pointer to point to n*, that is, determine n* as the next path node after S0. Thereafter, the current prediction node can be set to be added to the closed list, and f(pre) is recalculated. Here, calculating f(pre) is to calculate the path prediction cost of n*→pre + pre→Sg. Subsequently, the current prediction node is removed from the child node set, n* is moved from the open list to the closed list, the members in the child node set are put into the open list, and the binary tree structure corresponding to the open list is updated. Then, return to the step of determining whether the open list is empty.

[0081] Correspondingly, if g(new_pre) >= g(ole_pre), then directly jump to the step of removing the current prediction node from the child node set.

[0082] Execute the above process in a loop. When the destination node appears in the open list, it indicates that the reference planning path is obtained. If there is no data in the open list, it means that no suitable reference planning path is found. In the above process, the open list improved by the binary tree structure can quickly obtain the prediction node corresponding to the lowest path prediction cost each time during the path planning process, improving the efficiency of path planning.

[0083] It should be noted that as can be seen from the above, the core idea of the reverse path planning process is the same as that of the forward path planning process, that is, starting from adding the destination node to the open list and executing the path planning process, which will not be elaborated here.

[0084] In an exemplary embodiment, as Figure 5 shown, based on the path prediction cost of the prediction nodes in the path prediction region, using the path planning algorithm model to perform path planning from the forward and reverse directions respectively, so as to determine the reference planning path according to the forward and reverse planning results may include:

[0085] Step S510: determining a first prediction node corresponding to a minimum path prediction cost based on a first binary tree structure during the forward path planning process.

[0086] Step S520: Determine a second prediction node corresponding to the lowest path prediction cost based on the second binary tree structure during the reverse path planning process.

[0087] Among them, the open table of the forward path planning process corresponds to the first binary tree structure, and the open table of the reverse path planning process corresponds to the second binary tree structure. The execution order of step S510 and step S520 is not limited and can also be executed simultaneously.

[0088] Step S530: If the first prediction node is added to the closed list during the reverse path planning process, the planned path from the first prediction node to the starting node in the forward path planning process and the planned path from the destination node to the first prediction node in the reverse path planning process are spliced ​​to obtain a reference planned path; or, if the second prediction node is added to the closed list during the forward path planning process, the planned path from the destination node to the second prediction node in the reverse path planning process and the planned path from the starting node to the second prediction node in the forward path planning process are spliced ​​to obtain a reference planned path.

[0089] Specifically, in the bidirectional search strategy, traversing from the starting node to the destination node and from the destination node to the starting node, if the first predicted node with the lowest path prediction cost in the forward path planning process has been put into the closed list by the reverse search process, it means that the forward path planning has obtained the optimal path to the first predicted node, and the reverse path planning has also obtained the optimal path to the first predicted node. Similarly, if the second predicted node with the lowest path prediction cost in the reverse path planning process has been put into the closed list by the forward search process, it means that the reverse path planning has obtained the optimal path to the second predicted node, and the forward path planning has also obtained the optimal path to the second predicted node.

[0090] Based on this, through the first prediction node or the second prediction node, the parent node of each prediction node is tracked and connected to obtain a reference planning path, that is, the planning path from the first prediction node to the starting node in the forward path planning process is spliced ​​with the planning path from the destination node to the first prediction node in the reverse path planning process, or the planning path from the destination node to the second prediction node in the reverse path planning process is spliced ​​with the planning path from the starting node to the second prediction node in the forward path planning process to obtain a reference planning path. This process can effectively reduce the number of nodes searched in the algorithm process and further improve the efficiency of path planning.

[0091] In an exemplary embodiment, if the forward and reverse planning results obtained through the forward path planning process and the reverse path planning process, respectively, do not contain overlapping prediction nodes, the forward planning result or the reverse planning result is determined as the reference planning path.

[0092] Specifically, the first prediction node corresponding to the minimum path prediction cost determined based on the first binary tree structure in the forward path planning process and the second prediction node corresponding to the minimum path prediction cost determined based on the second binary tree structure in the reverse path planning process never overlap, that is, the forward path planning and the reverse path planning respectively obtain their own path planning results, so that the forward planning result or the reverse planning result can be determined as the reference planning path.

[0093] Among them, the total path prediction cost corresponding to the forward planning result can be compared with the total path prediction cost corresponding to the reverse planning result, and the forward planning result or the reverse planning result can be determined as the reference planning path according to the comparison result. The total path prediction cost of a path can be the sum of the path prediction costs F(S0→node+node→Sg) corresponding to each node in the path, or the total path prediction cost of the path is the sum of the path costs of two adjacent nodes, such as the path cost of S0→S1+the path cost of S1→S2+…+the path cost of Sn-1→Sn+the path cost of Sn→Sg, where n is the passing node in the reference planning path except the starting node and the destination node. Of course, the exemplary embodiments of the present disclosure can also determine the total path prediction cost of the path by other calculation methods, including but not limited to the above-mentioned calculation method.

[0094] By comparing the total path prediction costs corresponding to the forward and reverse planning results, the optimal path with the best path cost can be determined as the reference planning path, thereby improving the accuracy of path planning.

[0095] In an exemplary embodiment, a method for implementing path smoothing is also provided. Performing path smoothing on a reference planned path to obtain a target planned path may include:

[0096] If the number of nodes in the reference planning path is greater than the preset number threshold, the first smoothing strategy is used to eliminate redundant nodes in the reference planning path, and the reference planning path after eliminating redundant nodes is smoothed by the second smoothing strategy; if the number of nodes in the reference planning path is less than or equal to the preset number threshold, the second smoothing strategy is used to smooth the reference planning path.

[0097] Specifically, the first smoothing strategy is used to reduce the number of turning points of the reference planning path, and the second smoothing strategy is used to remove nodes based on the distances of adjacent nodes. In other words, the number of nodes in the reference planning path can be reduced by the first smoothing strategy and the second smoothing strategy.

[0098] Among them, the prerequisite for using the first smoothing strategy to eliminate redundant nodes in the reference planning path is that the reference planning path includes at least a preset number threshold number of nodes (such as at least four nodes). Therefore, it is first determined whether the number of nodes in the reference planning path is greater than the preset number threshold. If so, the reference planning path is smoothed by the first smoothing strategy, otherwise the reference planning path is directly processed by the second smoothing strategy to eliminate nodes. Of course, after the reference planning path is processed by the first smoothing strategy, the node elimination process can also be continued through the second smoothing path.

[0099] In an exemplary embodiment, using a first smoothing strategy to remove redundant nodes in a reference planning path includes:

[0100] The following process is repeated in the order from the starting node to the destination node until the number of nodes in the reference planning path meets the preset number of nodes:

[0101] Obtain a ternary continuous node group of the current node to be checked from the reference planning path;

[0102] If the ternary continuous node group satisfies the node elimination strategy, the second node in the ternary continuous node group is eliminated, the reference planning path is updated according to the elimination result, and the process returns to the step of obtaining the ternary continuous node group of the current node to be checked from the reference planning path;

[0103] If the ternary continuous node group does not satisfy the node elimination strategy, the current node to be checked is updated based on the reference planning path, and the process returns to the step of obtaining the ternary continuous node group of the current node to be checked from the reference planning path.

[0104] Among them, the ternary continuous node group refers to taking the current detection node as the first node, based on the reference planning path, obtaining the node adjacent to the first node as the second node, and obtaining the node adjacent to the second node as the third node. For example, for the reference planning path A(S0)-BCDEFG(Sg), if A is the current detection node, a ternary continuous node group is formed according to nodes A, B and C, and then judging whether the ternary continuous node group satisfies the node elimination strategy.

[0105] Specifically, the node elimination strategy includes: determining that the third node in the ternary continuous node group is directly visible relative to the first node according to the position information of each node in the ternary continuous node group; and the distance between the first node and the third node is less than or equal to the sum of the distance between the first node and the second node and the distance between the second node and the third node. If the ternary continuous node group satisfies the node elimination strategy, node B is eliminated from the reference planning path, and the new path becomes A(S0)-CDEFG(Sg), as shown in Figure 6As shown, node B can be removed. Then, the three-element continuous node group (node ​​A, node C and node D) can be continuously obtained and the above judgment process can be repeated. Based on this, the number of nodes on the reference planning path will decrease after each cycle until the nodes on the reference planning path are reduced to a feasible number (preset number of nodes).

[0106] Furthermore, there may still be redundant nodes in the reference planning path after being processed by the first smoothing strategy, such as Figure 6 As shown in FIG. 1 , the node C and the node D are found in actual implementation to exist at path points that may be associated with unfavorable vehicle tracking. To solve this problem, the reference planned path may be further processed through a second smoothing strategy.

[0107] Considering that the size of the grid cells in the grid map is clearly defined, in the navigation process that requires large direction changes, especially between two adjacent nodes that are relatively close to each other, compromised tracking results may be produced, which is mainly due to the limitation of vehicle calibration capabilities. Based on this, in order to compensate for the impact of tracking errors, the optimal path control distance (expressed as d cont Represented by), eliminating redundant nodes based on the optimal path control distance can be achieved through the following formula (1):

[0108] |d(X n+l )-d(X n )|<d cont (1)

[0109] Among them, if two adjacent nodes X n and X n-1 The distance between them is less than the optimal path control distance d cont , then delete the redundant point X n-1 .

[0110] It should be noted that the optimal path control distance can be determined based on the performance of a pre-completed experiment for identifying a dynamic model. In addition, the optimal path control distance is also based on actual experience and is related to the operating environment (which may be affected by fluid dynamics). The optimal path control distance can be determined based on the actual application scenario, and the illustrative examples of the present disclosure do not limit this.

[0111] Figure 7 A flow chart of path smoothing for a reference planning path is shown below. Figure 7 The path smoothing process is described.

[0112] First, it is determined whether the number of nodes of the reference planned path is greater than a preset number threshold.

[0113] If the number of nodes in the reference planning path is greater than a preset number threshold, the execution step of the first smoothing strategy is entered. Specifically, the current detection node is determined from the reference planning path in the order from the starting node to the destination node, and the first node selected as the current detection node is the first node in the reference planning path (ie, S0), and so on.

[0114] Then, the ternary continuous node group of the current node to be checked is obtained from the reference planning path, and the ternary continuous node group is defined as (A(S0), B, C).

[0115] Next, based on the position information of each node in the ternary continuous node group, it is determined whether the third node (C) in the ternary continuous node group is directly visible relative to the first node (A).

[0116] Among them, if the third node (C) is directly visible relative to the first node (A), it is determined whether the distance between the first node (A) and the third node (C) is less than or equal to the sum of the distance between the first node (A) and the second node (B) and the distance between the second node (B) and the third node (C). If so, the second node (B) is deleted from the reference planning path, and the reference planning path is updated. In addition, it is determined whether the subsequent adjacent node of the third node is the destination node. If so, the process of path smoothing using the first smoothing strategy is terminated. If not, jump to the step of obtaining the ternary continuous node group of the current check node from the reference planning path.

[0117] If it is determined that the third node (C) in the ternary continuous node group is not directly visible relative to the first node (A), or the distance between the first node (A) and the third node (C) is greater than the sum of the distance between the first node (A) and the second node (B) and the distance between the second node (B) and the third node (C), then the current check node is set to be updated to B, that is, node A is set = node B, and return to the step of obtaining the ternary continuous node group of the current node to be checked from the reference planning path.

[0118] Among them, when returning to the step of obtaining the ternary continuous node group of the current node to be checked from the reference planning path, set node B = node C, and set node C = the subsequent adjacent node of node C on the reference planning path.

[0119] After the above process, the reference planning path is smoothed using the first smoothing strategy. Figure 7 As shown, using the second smoothing strategy to smooth the reference planning path may include:

[0120] First, it can be determined whether the number of turns of the reference planned path meets the preset turning condition. If so, the smoothing process is continued through the second smoothing strategy, otherwise it returns to the process of processing using the first smoothing strategy.

[0121] Secondly, based on the reference planning path entering the second smoothing strategy processing flow, the ternary continuous node group (A, B, C) is re-determined.

[0122] Then, determine whether the distance between node A and node B in the three-element continuous node group is less than the optimal path control distance d cont If yes, node B is removed from the reference planning path, and it is determined whether the subsequent adjacent node of the third node (C) is the destination node. If yes, the process of path smoothing using the second smoothing strategy is terminated. If no, the process jumps to the step of obtaining a ternary continuous node group from the reference planning path, that is, setting node B = node C, and setting node C = the subsequent adjacent node of node C on the reference planning path, and then repeating the above process.

[0123] Among them, if the distance between node A and node B in the three-element continuous node group is not less than the optimal path control distance d cont , then directly jump to the step of determining whether the subsequent adjacent node of the third node (C) is the destination node, without eliminating node B.

[0124] The reference planning path is smoothed successively by the first smoothing strategy and the second smoothing strategy, thereby reducing the number of turning points and the redundant nodes in the path.

[0125] In an exemplary embodiment, the reference planning path processed by the first smoothing strategy and the second smoothing strategy is a node connected by a series of rigid straight lines. To further improve the smooth continuity of the path, the following may also be included:

[0126] Interpolation processing is performed on adjacent nodes in the reference planning path after smoothing by the second smoothing strategy to obtain continuous path points, and the obtained continuous path points are connected to obtain the target planning path.

[0127] Among them, the nodes on the reference planning path can be used to interpolate a series of points for smoothing the path using the spline interpolation method. For example, the reference planning path can be smoothed using the cubic spline interpolation method. Among them, the main principle of cubic spline interpolation is to use a spline composed of a third-order polynomial to connect a set of control points (or nodes on the path). The interval between two control points (nodes) has the form of a polynomial function, defined as:

[0128] f k (x) = A 3k (xx k ) 3 +A 2k (xx k ) 2 +A 1k(xx k )+A 0k (2)

[0129] Where k is the index of the node on the reference planning path. By giving the coordinate value of each node, the coefficients of the polynomial are calculated using constraints (A 0k ,A 1k ,A 2k ,A 3k ).

[0130] Constraints may include: a cubic polynomial that matches the equation value at both ends of the interval (between two nodes) [x k ,x k+1 ], that is: f k (x k )=y k and f k (x k+1 )=y k+1 , the first-order (velocity) and second-order (acceleration) derivatives of the cubic spline function at each interval node have the same value, making the generated path continuous, that is, f' k-1 (x k ) = f' k (x k ) and f” k-1 (x k )=f” k (x k ), the first-order derivatives of the starting node and the destination node are equal to the initial velocity and the final velocity, and the second-order derivatives of the starting node and the destination node are zero.

[0131] Based on the cubic spline interpolation method, we can first calculate the polynomial coefficients (A 0k ,A 1k ,A 2k ,A 3k ), then interpolate between two nodes and calculate the spline value, and finally obtain a smooth path. The cubic spline interpolation method can be found in the relevant technology and will not be described here.

[0132] By continuing to interpolate the reference planning path, a smooth and continuous target planning path is obtained, which further improves the effect of the path planning result.

[0133] It should be noted that the exemplary embodiments of the present disclosure may also select other path smoothing strategies to smooth the reference planned path according to actual needs, such as the Floyd algorithm, the PRM (Probabilistic Roadmap Method) algorithm, the QPMI (Quadratic Polynomial and Membership Interpolation) algorithm, etc., without limitation to this.

[0134] In the exemplary embodiment of the present disclosure, the path planning method, on the one hand, obtains the path prediction area including the starting node and the destination node, and constructs the path planning algorithm model. The path planning algorithm model is obtained by improving the open list of the heuristic search algorithm model based on the binary tree structure, so that the root node of the binary tree structure is set to the prediction node corresponding to the lowest path prediction cost in the open list. The path planning algorithm model based on the improved heuristic search algorithm model only needs to target the root node of the binary tree when obtaining and deleting nodes during path planning. Based on the binary tree structure, the time complexity required for node comparison can be reduced, and the path planning efficiency can be improved. On the other hand, the path planning algorithm model is used to perform path planning from the forward and reverse directions respectively, so as to determine the reference planning path according to the forward and reverse planning results. Based on the two-way search strategy, the optimal path can be found faster, and the planning efficiency of the path can be further improved. On the other hand, the target planning path obtained by path smoothing the reference planning path can eliminate the "sawtooth" path existing in the planning result based on the path prediction cost during the path planning process, reduce the turning cost of the automatic driving device, avoid the situation of path discontinuity, and thus avoid the problem of motion instability and overdriving caused by the path non-smoothness.

[0135] In an exemplary embodiment of the present disclosure, a path planning device is also provided. Figure 8 As shown, the path planning device 800 may include an area acquisition module 810, an algorithm construction module 820, a path planning module 830 and a path smoothing module 840. Specifically:

[0136] The area acquisition module 810 is used to obtain a path prediction area including a start node and a destination node, and the path prediction area also includes multiple prediction nodes; the algorithm construction module 820 is used to construct a path planning algorithm model, and the path planning algorithm model is obtained by improving the open list of the heuristic search algorithm model based on a binary tree structure, and the root node of the binary tree structure is set to the prediction node corresponding to the lowest path prediction cost in the open list; the path planning module 830 is used to perform path planning from forward and reverse directions respectively based on the path prediction costs of the prediction nodes in the path prediction area using the path planning algorithm model, so as to determine a reference planning path according to the forward and reverse planning results; the path smoothing module 840 is used to perform path smoothing on the reference planning path to obtain a target planning path.

[0137] In an exemplary embodiment of the present disclosure, the heuristic search algorithm model includes an open list for storing unpredicted nodes in the path planning process and a closed list for storing predicted nodes in the path planning process; the algorithm construction module 820 is configured to execute: determining the path prediction cost of the unpredicted nodes; constructing a binary tree structure based on the open list, so that the root node of the binary tree structure is always the predicted node corresponding to the lowest path prediction cost in the open list, and the path prediction cost of the child node is greater than or equal to the path prediction cost of the corresponding parent node.

[0138] In an exemplary embodiment of the present disclosure, the path planning module 830 is configured to execute:

[0139] Based on the path prediction cost of the prediction node in the path prediction area, the forward path planning is performed from the starting node to the destination node using the path planning algorithm model, and the reverse path planning is performed from the destination node to the starting node using the path planning algorithm model; wherein the path planning algorithm model includes a first binary tree structure corresponding to the forward path planning, a second binary tree structure corresponding to the reverse path planning, and a closed list corresponding to the forward path planning and the reverse path planning.

[0140] In an exemplary embodiment of the present disclosure, the path planning module 830 is configured to execute:

[0141] In the forward path planning process, the first prediction node corresponding to the lowest path prediction cost is determined based on the first binary tree structure; in the reverse path planning process, the second prediction node corresponding to the lowest path prediction cost is determined based on the second binary tree structure; if the first prediction node is added to the closed list in the reverse path planning process, the planned path from the first prediction node to the starting node in the forward path planning process is spliced ​​with the planned path from the destination node to the first prediction node in the reverse path planning process to obtain a reference planned path; or, if the second prediction node is added to the closed list in the forward path planning process, the planned path from the destination node to the second prediction node in the reverse path planning process is spliced ​​with the planned path from the starting node to the second prediction node in the forward path planning process to obtain a reference planned path.

[0142] In an exemplary embodiment of the present disclosure, the path planning module 830 is further configured to execute:

[0143] If the forward and reverse planning results obtained through the forward path planning process and the reverse path planning process respectively do not contain overlapping prediction nodes, the forward planning result or the reverse planning result is determined as the reference planning path.

[0144] In an exemplary embodiment of the present disclosure, the path planning module 830 is configured to execute:

[0145] The total predicted path cost corresponding to the forward planning result is compared with the total predicted path cost corresponding to the reverse planning result; and the forward planning result or the reverse planning result is determined as a reference planning path according to the comparison result.

[0146] In an exemplary embodiment of the present disclosure, the path smoothing module 840 is configured to perform:

[0147] If the number of nodes in the reference planning path is greater than a preset number threshold, the first smoothing strategy is used to eliminate redundant nodes in the reference planning path, and the reference planning path after eliminating redundant nodes is smoothed by the second smoothing strategy; if the number of nodes in the reference planning path is less than or equal to the preset number threshold, the second smoothing strategy is used to smooth the reference planning path; wherein the second smoothing strategy is used to eliminate nodes based on the distance between adjacent nodes.

[0148] In an exemplary embodiment of the present disclosure, the path smoothing module 840 is configured to perform:

[0149] The following process is executed in a loop from the starting node to the destination node until the number of nodes in the reference planning path meets the preset number of nodes: a ternary continuous node group of the current node to be checked is obtained from the reference planning path; if the ternary continuous node group meets the node elimination strategy, the second node in the ternary continuous node group is eliminated, the reference planning path is updated according to the elimination result, and the process returns to the step of obtaining the ternary continuous node group of the current node to be checked from the reference planning path; if the ternary continuous node group does not meet the node elimination strategy, the current node to be checked is updated based on the reference planning path, and the process returns to the step of obtaining the ternary continuous node group of the current node to be checked from the reference planning path.

[0150] In an exemplary embodiment of the present disclosure, a node removal strategy includes:

[0151] According to the position information of each node in the ternary continuous node group, it is determined that the third node in the ternary continuous node group is directly visible relative to the first node; and the distance between the first node and the third node is less than or equal to the sum of the distance between the first node and the second node and the distance between the second node and the third node.

[0152] In an exemplary embodiment of the present disclosure, the path smoothing module 840 is further configured to perform:

[0153] Interpolation processing is performed on adjacent nodes in the reference planning path after smoothing by the second smoothing strategy to obtain continuous path points; the obtained continuous path points are connected to obtain the target planning path.

[0154] Since the details of the various functional modules of the path planning device of the exemplary embodiment of the present disclosure have been recorded in the exemplary embodiment of the path planning method described above, they will not be repeated here.

[0155] It should be noted that, although several modules or units of the path planning device are mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided into multiple modules or units to be embodied.

[0156] The exemplary embodiments of the present disclosure also provide a computer program product, which includes a computer program, and when the computer program is executed by a processor, the above-mentioned path planning method is implemented.

[0157] In one embodiment, the computer program product may be a tangible product containing a computer program, such as a computer-readable storage medium storing a computer program. The readable storage medium may be a storage medium based on electrical, magnetic, optical, electromagnetic, infrared, or other signals, including but not limited to: random access memory (RAM), read-only memory (ROM), magnetic tape, floppy disk, flash memory (Flash), mechanical hard disk (HDD), solid-state drive (SSD), and the like. Exemplarily, the computer program product may be implemented as a non-volatile storage medium storing a computer program, such as a read-only memory, a NAND flash memory, and the like.

[0158] In one embodiment, the computer program product may be an intangible product including a computer program. Exemplarily, the computer program product may be implemented as a virtual digital product, such as a digital file storing an executable file, an installation package, etc. of the computer program.

[0159] The code of the computer program can be written in one or more programming languages. Programming languages ​​such as C language, Java, C++, etc. The program code can be executed entirely on the user computing device, or partially on the user computing device, or as a separate software package, or partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device can be connected to the user computing device through any type of network, such as a local area network (LAN), a wide area network (WAN), etc., or can be connected to an external computing device (e.g., an Internet connection provided by an operator).

[0160] The computer program may be carried or transmitted via electrical, magnetic, optical, electromagnetic, infrared, or other signals. The electronic device may convert the signal carrying the computer program into a digital signal, thereby running the computer program. When the computer program is run on an electronic device, its code is used to enable the electronic device to execute (more specifically, the processor of the electronic device may execute) the method steps of various exemplary embodiments of the present disclosure, such as the path planning method described above.

[0161] In addition, in an exemplary embodiment of the present disclosure, an electronic device capable of implementing the above method is also provided. It will be appreciated by those skilled in the art that various aspects of the present disclosure may be implemented as a system, method, or program product. Therefore, various aspects of the present disclosure may be specifically implemented in the following forms, namely: a complete hardware embodiment, a complete software embodiment (including firmware, microcode, etc.), or an embodiment combining hardware and software aspects, which may be collectively referred to herein as a "circuit", "module", or "system".

[0162] Refer to the following Fig. 9hereinafter describes an electronic device 900 according to such an embodiment of the present disclosure. Fig. 9 The electronic device 900 shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0163] like Fig. 9 As shown, the electronic device 900 is in the form of a general computing device. The components of the electronic device 900 may include, but are not limited to: the at least one processing unit 910, the at least one storage unit 920, a bus 930 connecting different system components (including the storage unit 920 and the processing unit 910), and a display unit 940.

[0164] The storage unit stores a program code, which can be executed by the processing unit 910, so that the processing unit 910 performs the steps according to various exemplary embodiments of the present disclosure described in the above “Exemplary Method” section of this specification. For example, the processing unit 910 can perform the steps shown in the above exemplary embodiments.

[0165] The storage unit 920 may include a readable medium in the form of a volatile storage unit, such as a random access memory unit (RAM) 921 and / or a cache memory unit 922 , and may further include a read-only memory unit (ROM) 923 .

[0166] The storage unit 920 may also include a program / utility 924 having a set (at least one) of program modules 925, such program modules 925 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.

[0167] Bus 930 may represent one or more of several types of bus structures, including a memory unit bus or memory unit controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.

[0168] The electronic device 900 may also communicate with one or more external devices 1000 (e.g., keyboards, pointing devices, Bluetooth devices, etc.), may also communicate with one or more devices that enable a user to interact with the electronic device 900, and / or communicate with any device that enables the electronic device 900 to communicate with one or more other computing devices (e.g., routers, modems, etc.). Such communication may be performed via an input / output (I / O) interface 950. Furthermore, the electronic device 900 may also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) via a network adapter 960. As shown, the network adapter 960 communicates with other modules of the electronic device 900 via a bus 930. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 900, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, etc.

[0169] Through the description of the above embodiments, it is easy for those skilled in the art to understand that the example embodiments described here can be implemented by software, or by software combined with necessary hardware. Therefore, the technical solution according to the embodiment of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, and includes several instructions to enable a computing device (which can be a personal computer, a server, a terminal device, or a network device, etc.) to execute the method according to the embodiment of the present disclosure.

[0170] In addition, the above-mentioned figures are only schematic illustrations of the processes included in the method according to the exemplary embodiments of the present disclosure, and are not intended to be limiting. It is easy to understand that the processes shown in the above-mentioned figures do not indicate or limit the time sequence of these processes. In addition, it is also easy to understand that these processes can be performed synchronously or asynchronously, for example, in multiple modules.

[0171] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. The present disclosure is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary technical means in the art that are not disclosed in the present disclosure. The description and examples are to be considered exemplary only, and the true scope and spirit of the present disclosure are indicated by the claims.

Claims

1. A path planning method, characterized in that: include: Acquire a path prediction area including a start node and a destination node, wherein the path prediction area also includes a plurality of prediction nodes; Constructing a path planning algorithm model, wherein the path planning algorithm model is obtained by improving an open list of a heuristic search algorithm model based on a binary tree structure, wherein the root node of the binary tree structure is set to a prediction node corresponding to a lowest path prediction cost in the open list; Based on the path prediction costs of the prediction nodes in the path prediction area, the path planning algorithm model is used to perform path planning from a forward direction and a reverse direction respectively, so as to determine a reference planning path according to the forward and reverse planning results; The reference planned path is smoothed to obtain a target planned path.

2. The method according to claim 1, characterized in that The heuristic search algorithm model includes an open list for storing unpredicted nodes in the path planning process and a closed list for storing predicted nodes in the path planning process; The construction of the path planning algorithm model includes: Determining a path prediction cost of the unpredicted node; The binary tree structure is constructed according to the open list, so that the root node of the binary tree structure is always the prediction node corresponding to the lowest path prediction cost in the open list, and the path prediction cost of the child node is greater than or equal to the path prediction cost of the corresponding parent node.

3. The method according to claim 1, characterized in that The path prediction cost based on the prediction node in the path prediction area, using the path planning algorithm model to perform path planning from forward and reverse directions respectively, so as to determine a reference planning path according to forward and reverse planning results, includes: Based on the path prediction costs of the prediction nodes in the path prediction area, forward path planning is performed from the starting node to the destination node using the path planning algorithm model, and reverse path planning is performed from the destination node to the starting node using the path planning algorithm model; The path planning algorithm model includes a first binary tree structure corresponding to the forward path planning, a second binary tree structure corresponding to the reverse path planning, and a closed list table corresponding to the forward path planning and the reverse path planning.

4. The method according to claim 3, characterized in that: The path prediction cost based on the prediction node in the path prediction area, using the path planning algorithm model to perform path planning from forward and reverse directions respectively, so as to determine a reference planning path according to forward and reverse planning results, includes: Determining a first prediction node corresponding to a minimum path prediction cost based on the first binary tree structure during the forward path planning process; Determining a second prediction node corresponding to a minimum path prediction cost based on the second binary tree structure during the reverse path planning process; If the first prediction node is added to the closed list in the reverse path planning process, the planned path from the first prediction node to the start node in the forward path planning process is concatenated with the planned path from the destination node to the first prediction node in the reverse path planning process to obtain the reference planned path; Alternatively, if the second prediction node is added to the closed list during the forward path planning process, the planned path from the destination node to the second prediction node during the reverse path planning process and the planned path from the starting node to the second prediction node during the forward path planning process are spliced ​​to obtain the reference planned path.

5. The method according to claim 4, characterized in that The method further comprises: If the forward and reverse planning results obtained through the forward path planning process and the reverse path planning process respectively do not contain overlapping prediction nodes, the forward planning result or the reverse planning result is determined as the reference planning path.

6. The method according to claim 5, characterized in that The determining the forward planning result or the reverse planning result as the reference planning path includes: Comparing the total path prediction cost corresponding to the forward planning result with the total path prediction cost corresponding to the reverse planning result; The forward planning result or the reverse planning result is determined as the reference planning path according to the comparison result.

7. The method according to claim 1, characterized in that The step of performing path smoothing on the reference planned path to obtain a target planned path includes: If the number of nodes in the reference planning path is greater than a preset number threshold, a first smoothing strategy is used to remove redundant nodes in the reference planning path, and a second smoothing strategy is used to smooth the reference planning path after removing the redundant nodes; If the number of nodes of the reference planned path is less than or equal to the preset number threshold, the second smoothing strategy is used to smooth the reference planned path; The second smoothing strategy is used to remove nodes based on the distance between adjacent nodes.

8. The method according to claim 7, characterized in that The adopting the first smoothing strategy to eliminate redundant nodes in the reference planning path includes: The following process is executed in a loop from the starting node to the destination node until the number of nodes in the reference planning path meets the preset number of nodes: Acquire a ternary continuous node group of a current node to be checked from the reference planning path; If the ternary continuous node group satisfies the node elimination strategy, the second node in the ternary continuous node group is eliminated, the reference planning path is updated according to the elimination result, and the step of obtaining the ternary continuous node group of the current node to be checked from the reference planning path is returned; If the ternary continuous node group does not satisfy the node elimination strategy, the current node to be checked is updated based on the reference planning path, and the process returns to the step of obtaining the ternary continuous node group of the current node to be checked from the reference planning path.

9. The method according to claim 8, characterized in that The node removal strategy includes: Determining, based on position information of each node in the ternary consecutive node group, that a third node in the ternary consecutive node group is directly visible relative to the first node; and The distance between the first node and the third node is less than or equal to the sum of the distance between the first node and the second node and the distance between the second node and the third node.

10. The method according to claim 7, characterized in that For the reference planning path after smoothing by the second smoothing strategy, it also includes: Interpolating adjacent nodes in the reference planning path after smoothing by the second smoothing strategy to obtain continuous path points; The obtained continuous path points are connected to obtain the target planning path.

11. A path planning device, characterized in that: The device comprises: An area acquisition module, used to acquire a path prediction area including a start node and a destination node, wherein the path prediction area also includes a plurality of prediction nodes; An algorithm construction module, used to construct a path planning algorithm model, wherein the path planning algorithm model is obtained by improving the open list of the heuristic search algorithm model based on a binary tree structure, and the root node of the binary tree structure is set to the prediction node corresponding to the lowest path prediction cost in the open list; A path planning module, configured to perform path planning in a forward direction and a reverse direction respectively based on the path prediction costs of the prediction nodes in the path prediction area using the path planning algorithm model, so as to determine a reference planning path according to the forward and reverse planning results; The path smoothing module is used to perform path smoothing processing on the reference planning path to obtain a target planning path.

12. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method according to any one of claims 1 to 10 is implemented.

13. An electronic device, characterized in that: include: processor; as well as A memory, configured to store executable instructions of the processor; The processor is configured to perform the method of any one of claims 1 to 10 by executing the executable instructions.

Images