Mobile robot path planning method and system

Abstract

The invention relates to a mobile robot path planning method and system, and the method comprises the steps: determining a multi-neighborhood mixed search mechanism for the mixed use of eight neighborhoods and new sixteen neighborhoods according to whether there is an obstacle in the eight neighborhoods of a current node; deleting invalid nodes according to the positions of the current node and the target node; a bidirectional search strategy is adopted, a dynamic weight coefficient is added to adjust the heuristic function, and a planned path is obtained; and a redundant inflection point deletion strategy is adopted for the planned path, useless inflection points are removed, and an optimal path is obtained. By improving a neighborhood search mechanism, the search efficiency of a path planning algorithm is improved; through a node optimization strategy, invalid nodes in the algorithm searching process are removed, and the path planning algorithm searching time is shortened; by introducing a two-way search mechanism, the algorithm search efficiency is improved, meanwhile, the dynamic weight factor is added to adjust the algorithm search speed, search of invalid regions is reduced, and therefore the search efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of path planning, and in particular to a mobile robot path planning method and system. Background Art

[0002] Path planning has always been one of the key technologies in the field of mobile robot research. The main purpose of path planning is to quickly and accurately find an optimal path from the starting position to the target position without collision in an unknown environment.

[0003] At present, researchers have developed a variety of path planning algorithms to cope with different environments and needs. The path planning of mobile robots can be divided into three categories: traditional algorithms, sampling-based planning methods, and intelligent algorithms. Traditional algorithms such as Dijkstra algorithm and A* algorithm are classic algorithms in path planning. The A* algorithm improves the search efficiency by introducing heuristic functions, but it cannot guarantee the global optimal solution; sampling-based planning methods include probabilistic road map method (PRM) and rapid search random tree method (RRT), which explore the state space through random sampling and construct feasible paths. They are particularly suitable for problems with high-dimensional space and complex constraints, but may require a large number of samples to ensure the feasibility of the path; the application of intelligent algorithms is becoming more and more extensive. With the development of artificial intelligence technology, more and more intelligent algorithms are being applied to path planning, such as neural networks, fuzzy logic, genetic algorithms, etc. These algorithms can provide more flexible and adaptable solutions, especially in dynamic and unknown environments. In summary, the research on mobile robot path planning is moving towards improving adaptability and efficiency, as well as smarter decision-making. With the continuous advancement of technology, future path planning algorithms will be more intelligent and automated, and will be able to better adapt to complex and changing environments.

[0004] The A* algorithm is a heuristic search algorithm that is widely used in the field of path planning, including path planning for mobile robots. Although the A* algorithm is favored for its efficiency and accuracy, it also has some disadvantages in practical applications: 1) Slow search speed: When using the A* algorithm for mobile robot path planning, the commonly used neighborhood search methods are four-neighborhood and eight-neighborhood search modes. When the search space is large and there are few obstacles, the efficiency of this search method will be very low, resulting in a slow search speed of the algorithm. 2) The existence of useless nodes: When the A* algorithm expands outward, it needs to calculate the cost function of all possible next positions. However, in the actual search process, the nodes that deviate from the search direction of the final target position are often invalid nodes, which will take up the algorithm search time and reduce the algorithm search efficiency. 3) Heuristic function dependency: The performance of the A* algorithm depends greatly on the choice of heuristic function. If the heuristic function is not selected properly, it may lead to reduced algorithm efficiency or failure to find the optimal solution. 4) The existence of redundant nodes: The path obtained using the A* algorithm often has many redundant nodes, resulting in more inflection points in the path, which reduces the efficiency of the mobile robot.

[0005] Existing mobile robot path planning algorithms usually have problems such as low search efficiency, too many extended nodes, and redundant nodes. In the prior art, the invention with application number 202110837144.2 discloses the use of the RRT* algorithm to generate multiple feasible paths, and calculates the time cost of each path through the turning characteristic function, and finally selects the path with the smallest time cost as the planning path. This method can reduce the path time cost of the actual task of the robot and improve the efficiency of task execution, but the amount of calculation is large, resulting in low algorithm operation efficiency. The invention with application number 201711374451.1 discloses a path planning method, firstly performing global path pre-planning in a static obstacle environment, and then performing secondary smoothing on the pre-planned path. When encountering dynamic obstacles, the artificial potential field method is used for local path planning. This method effectively improves the efficiency and smoothness of path planning, but when faced with frequent changes in the dynamic environment, the real-time update and adaptability of the algorithm are insufficient. The application number is invention 202010639974.X, which discloses the use of the A* algorithm for global path planning and the use of weighted ideas to improve the evaluation function to eliminate redundant points. When encountering dynamic obstacles, the artificial potential field algorithm is used for local path planning. However, this method increases the computational complexity and reduces the computational efficiency of the algorithm through algorithm fusion.

[0006] It can be seen that the existing technology still has many defects in mobile robot path planning, and it is urgent to propose a new path planning method to improve the overall efficiency of robot path planning. Summary of the invention

[0007] In view of this, it is necessary to provide a mobile robot path planning method and system to solve the above-mentioned defects of the prior art.

[0008] In order to solve the above problems, in a first aspect, an embodiment of the present invention provides a mobile robot path planning method, comprising:

[0009] S1, according to whether there are obstacles in the eight neighborhoods of the current node, determine the multi-neighborhood hybrid search mechanism using the eight neighborhoods and the new sixteen neighborhoods;

[0010] S2, based on the multi-neighborhood hybrid search mechanism, delete invalid nodes according to the positions of the current node and the target node;

[0011] S3, adopts a bidirectional search strategy and adds a dynamic weight coefficient adjustment heuristic function to obtain the planned path;

[0012] S4, adopt the redundant inflection point deletion strategy for the planned path, remove useless inflection points and obtain the optimal path.

[0013] Preferably, before step S1, the method further comprises:

[0014] The grid method is used to simulate the environment and obtain a grid map for mobile robot path planning.

[0015] Preferably, in step S1, according to whether there are obstacles in the eight neighborhoods of the current node, a multi-neighborhood hybrid search mechanism using eight neighborhoods and new sixteen neighborhoods in combination is determined, including:

[0016] Determine whether there are obstacles in the eight neighborhoods of the current node:

[0017] If there are obstacles, the eight-neighborhood search method is used;

[0018] If there are no obstacles, a sixteen-neighborhood search method is used; wherein the sixteen-neighborhood search method is optimized based on an eight-neighborhood search method and a twenty-four-neighborhood search method.

[0019] Preferably, in step S2, based on the multi-neighborhood hybrid search mechanism, invalid nodes are deleted according to the positions of the current node and the target node, including:

[0020] Determine the positions of the current node and the target node in the grid map, and calculate the angle α between the line connecting the current node and the target node and the positive direction of the x-axis of the grid map;

[0021] According to the range of angle α and the current neighborhood search method, invalid nodes are removed: when using the eight-neighborhood search method, according to the range of angle α, 3 invalid nodes in the eight-neighborhood are removed, and 5 valid nodes are retained. When using the sixteen-neighborhood search method, according to the range of angle α, 9 invalid nodes in the sixteen-neighborhood are removed, and 7 valid nodes are retained.

[0022] Preferably, in step S3, a bidirectional search strategy is adopted, and a dynamic weight coefficient adjustment heuristic function is added to obtain a planned path, specifically including:

[0023] Initialize the forward search and reverse search lists; forward search searches from the starting point to the end point, and reverse search searches from the end point to the starting point;

[0024] Calculate the cost value of each node in the list according to the cost calculation function, and select the node with the smallest cost value as the node to be explored next;

[0025] Determine whether the forward search and the reverse search intersect. If so, the algorithm ends, otherwise, continue searching;

[0026] During the bidirectional search process, the dynamic weight coefficient is used to adjust the heuristic function to dynamically adjust the search speed.

[0027] Preferably, in the bidirectional search process, the heuristic function is adjusted using a dynamic weight coefficient to dynamically adjust the search speed, including:

[0028] In the bidirectional search process, the dynamic weight coefficient w(n) is used to adjust the heuristic function h(n) to dynamically balance the proportional relationship between g(n) and h(n); where g(n) represents the cost from the starting point to the current node n, and h(n) represents the cost from the current node n to the end point;

[0029] The calculation formula for the total cost of the bidirectional search algorithm after using the dynamic weight coefficient w(n) is:

[0030] f(n)=g(n)+w(n)·h(n)

[0031] In the formula, f(n) represents the total cost from the starting point to the end point;

[0032] Among them, in the forward search, the expression of the dynamic weight coefficient is:

[0033]

[0034] During reverse search, the expression of the dynamic weight coefficient is:

[0035]

[0036] In the formula, x sand s is the starting point coordinate, x g and g is the end point coordinate, x t and t is the current coordinate of the forward search, x u and u The current coordinates for reverse search;

[0037] The dynamic weight coefficient w(n) adopts a combination of exponential and logarithmic functions to dynamically adjust the search speed according to the distance between the current node and the target endpoint; if the distance between the current node and the target endpoint is greater than the preset distance, the search speed is accelerated; if the distance between the current node and the target endpoint is less than the preset distance, the search speed is reduced to improve the search accuracy.

[0038] Preferably, in step S4, the step of adopting a redundant inflection point deletion strategy for the planned path to remove useless inflection points includes:

[0039] S41, connecting the start node forward to the inflection points separated therefrom, and determining whether there is an obstacle in the line connecting the start node and the inflection points separated therefrom; wherein the end point of the planned path is taken as the initial start node;

[0040] S42, if there is no obstacle, delete the inflection point between the start node and the separated inflection point, and return to S41; if there is an obstacle, take the separated inflection point as an indelible inflection point, take the obtained indelible inflection point as a new start node, and return to S41;

[0041] S43, repeating S41 to S42, and gradually processing forward to the starting point of the path;

[0042] S44, reconnecting all the obtained indelible turning points to form an optimal path.

[0043] In a second aspect, an embodiment of the present invention provides a mobile robot path planning system, comprising:

[0044] A neighborhood search optimization module is used to determine a multi-neighborhood hybrid search mechanism using a mixture of eight neighborhoods and new sixteen neighborhoods according to whether there are obstacles in the eight neighborhoods of the current node;

[0045] A node optimization module, used to delete invalid nodes according to the positions of the current node and the target node based on the multi-neighborhood hybrid search mechanism;

[0046] A bidirectional search module is used to adopt a bidirectional search strategy and add a dynamic weight coefficient to adjust the heuristic function to obtain the planned path;

[0047] The redundant inflection point deletion module is used to adopt a redundant inflection point deletion strategy for the planned path, remove useless inflection points, and obtain the optimal path.

[0048] In a third aspect, the present invention further provides an electronic device, including a memory and a processor, wherein:

[0049] The memory is used to store programs;

[0050] The processor is coupled to the memory and is used to execute the program stored in the memory to implement the steps in the mobile robot path planning method as described in the embodiment of the first aspect of the present invention.

[0051] In a fourth aspect, the present invention further provides a computer-readable storage medium for storing computer-readable programs or instructions, which, when executed by a processor, can implement the steps in the mobile robot path planning method as described in the embodiment of the first aspect of the present invention.

[0052] Compared with the prior art, the mobile robot path planning method provided by the present invention has the following advantages:

[0053] Beneficial effects:

[0054] The mobile robot path planning method and system provided by the embodiment of the present invention improves the neighborhood search mechanism, expands the next target position that can be selected during the algorithm search, and improves the search efficiency of the path planning algorithm; removes invalid nodes in the algorithm search process through the node optimization strategy, and reduces the search time of the path planning algorithm; improves the efficiency of the algorithm search by introducing a two-way search mechanism, and at the same time adds a dynamic weight factor to adjust the algorithm search speed, ensuring that the two-way search can meet at the middle point and successfully plan the optimal path. Useless turning points are removed through the redundant turning point deletion strategy, which improves the efficiency of the mobile robot's travel. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 A flow chart of the mobile robot path planning method provided by the present invention;

[0056] Figure 2 A grid map for mobile robot path planning provided by the present invention;

[0057] FIG3( a) is a schematic diagram of the eight-neighborhood search provided by the present invention;

[0058] FIG3( b) is a schematic diagram of the twenty-four neighborhood search method provided by the present invention;

[0059] Figure 4 A schematic diagram of the sixteen-neighborhood search provided by the present invention;

[0060] FIG5( a) is a node diagram of the eight-neighborhood search provided by the present invention;

[0061] FIG5( b) is a node diagram of a sixteen-neighborhood search provided by the present invention;

[0062] Figure 6 A schematic diagram of redundant node deletion provided by the present invention;

[0063] Figure 7 A structural block diagram of the mobile robot path planning system provided by the present invention;

[0064] Figure 8 This is a structural block diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0065] The preferred embodiments of the present invention are described in detail below in conjunction with the accompanying drawings, wherein the accompanying drawings constitute a part of this application and are used together with the embodiments of the present invention to illustrate the principles of the present invention, but are not used to limit the scope of the present invention.

[0066] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0067] The A* algorithm is a heuristic search algorithm that is widely used in the field of path planning, including path planning for mobile robots. Although the A* algorithm is favored for its efficiency and accuracy, it also has some disadvantages in practical applications: 1) Slow search speed: When using the A* algorithm for mobile robot path planning, the commonly used neighborhood search methods are four-neighborhood and eight-neighborhood search modes. When the search space is large and there are few obstacles, the efficiency of this search method will be very low, resulting in a slow search speed of the algorithm. 2) The existence of useless nodes: When the A* algorithm expands outward, it needs to calculate the cost function of all possible next positions. However, in the actual search process, the nodes that deviate from the search direction of the final target position are often invalid nodes, which will take up the algorithm search time and reduce the algorithm search efficiency. 3) Heuristic function dependency: The performance of the A* algorithm depends greatly on the choice of heuristic function. If the heuristic function is not selected properly, it may lead to reduced algorithm efficiency or failure to find the optimal solution. 4) The existence of redundant nodes: The path obtained using the A* algorithm often has many redundant nodes, resulting in more inflection points in the path, which reduces the efficiency of the mobile robot.

[0068] In view of the above-mentioned defects of the traditional A* algorithm, an embodiment of the present invention provides a mobile robot path planning method, which improves the traditional A* algorithm and improves the overall efficiency of robot path planning. The following will be described and introduced through multiple embodiments.

[0069] Figure 1 This is a flow chart of the mobile robot path planning method provided by the present invention. Figure 1 As shown, the mobile robot path planning method includes steps S1 to S43, wherein:

[0070] Step S1, according to whether there are obstacles in the eight neighborhoods of the current node, determine a multi-neighborhood hybrid search mechanism that uses the eight neighborhoods and the new sixteen neighborhoods in combination.

[0071] Specifically, before executing step S1, the environment is simulated using the grid method to obtain a grid map for mobile robot path planning. Environmental modeling is one of the important links in the research of mobile robot path planning. A suitable map environment model helps to improve the search efficiency of the algorithm. The present invention uses the grid method to simulate environmental modeling. At the same time, in order to better realize the positioning of the mobile robot, and considering that the center point of the grid map is a regularly arranged point, the node of the path planning is limited to the center point of the grid, and the robot is regarded as a movable mass point, and its position is determined by the grid on the map, which can effectively reduce the complexity of the path.

[0072] Figure 2 A grid map for mobile robot path planning provided by the present invention. Figure 2 The 10x10 grid example is shown in the figure, where the grid in the lower left corner represents the starting position of the mobile robot, the grid in the upper right corner represents the end position of the mobile robot, the remaining black grids represent obstacles, and the white grids represent passable areas. After the grid map is established, each grid needs to be encoded. Each grid has only two states, idle or occupied. When the grid code is 1, that is, the grid color is black, it proves that the grid is occupied, that is, there is an obstacle; when the grid code is 0, that is, the grid color is white, it proves that the grid is idle, that is, the robot can pass normally.

[0073] After the grid map is established and the grid is encoded, step S1 is executed to optimize the neighborhood search method. Neighborhood search is to expand nodes outward from the current node in different directions. The traditional A* algorithm often uses an eight-neighborhood search method, that is, the neighborhood is expanded from the current node to 8 different directions, and the angle between each direction is 45°. The principle diagram of the eight-neighborhood search is shown in Figure 3(a). The coordinate set of the eight-neighborhood search is expressed as:

[0074] T n ={(xn ±1,y n ),(x n ,y n ±1),(x n ±1,y n ±1)} (1)

[0075] Where, T n is a set of coordinates, x n is the x-axis coordinate of the current search node n, y n is the y-axis coordinate of the current search node n.

[0076] Since the eight-neighborhood search method only expands to the surrounding eight nodes at a time, in an environment with fewer obstacles, there will be problems such as low search efficiency and too many expanded nodes. Therefore, a twenty-four-neighborhood search method is proposed. The twenty-four-neighborhood expansion method expands the search range from the first layer of nodes around the current node to the second layer of nodes. The algorithm can search towards 24 different nodes at a time. Its principle diagram is shown in Figure 3(b). The coordinate set of the twenty-four-neighborhood search method is expressed as:

[0077]

[0078] However, although the twenty-four neighborhood search method can explore multiple nodes at a time, its search efficiency will be greatly reduced when there are obstacles in the surrounding 1st layer nodes. When the algorithm expands to the 2nd layer nodes, it must first determine whether there are obstacles in the 1st layer nodes it passes through. At the same time, the multi-point search method will also cause the algorithm to explore too many useless nodes, greatly reducing the efficiency of the algorithm.

[0079] In view of the above-mentioned defects of the eight-neighborhood search method and the twenty-four-neighborhood search method, the embodiment of the present invention optimizes the eight-neighborhood search method and the twenty-four-neighborhood search method and proposes a new sixteen-neighborhood search method. The principle diagram of the sixteen-neighborhood search is as follows: Figure 4 The coordinate set of the sixteen-neighborhood search method is expressed as:

[0080]

[0081] The sixteen-neighborhood search method proposed in the present invention is based on the traditional eight-neighborhood and twenty-four-neighborhood expansion methods. The eight neighborhoods around the first-layer nodes are removed from the twenty-four neighborhoods to form a newly formed sixteen-neighborhood. When there are no obstacles in the eight neighborhoods of the first-layer nodes around the current node, the sixteen neighborhoods of the second-layer nodes can be directly explored.

[0082] In step S1, a multi-neighborhood hybrid search mechanism using a mixture of eight neighborhoods and new sixteen neighborhoods is determined, specifically: determine whether there are obstacles in the eight neighborhoods of the current node: if there are obstacles, use the eight-neighborhood search method; if there are no obstacles, use the sixteen-neighborhood search method to search. This multi-neighborhood hybrid search algorithm determines whether there are obstacles in the eight neighborhoods of the current node and determines whether to use the eight-neighborhood expansion method or the new sixteen-neighborhood expansion method. The algorithm can effectively perform cross-node exploration, effectively improve the algorithm's search efficiency, and reduce the expansion of useless nodes.

[0083] In the embodiment of the present invention, by improving the neighborhood search mechanism, the selectable next target position during the algorithm search is expanded, thereby improving the search efficiency of the path planning algorithm.

[0084] Step S2, based on the multi-neighborhood hybrid search mechanism, delete invalid nodes according to the positions of the current node and the target node.

[0085] Specifically, the present invention provides a multi-neighborhood hybrid search mechanism that uses a mixture of eight neighborhoods and new sixteen neighborhoods, and the search nodes are shown in Figures 5(a) and 5(b) respectively. In the process of exploring from the current node (the current location of the robot) to the target node (the location the robot needs to reach), invalid nodes will be generated when using eight neighborhoods or the new sixteen neighborhoods to search, which will waste resources for the algorithm search. In order to improve the efficiency of the algorithm search, the present invention removes useless nodes that have no effect on the search results during the search process.

[0086] First, determine the positions of the current node and the target node in the grid map, and calculate the angle α between the line connecting the current node and the target node and the positive direction of the x-axis of the grid map;

[0087] Then, invalid nodes are removed according to the range of the angle α and the current neighborhood search method.

[0088] When the eight-neighborhood search method is used, as shown in Table 1, according to the range of the angle α, three invalid nodes in the eight-neighborhood are removed and five valid nodes are retained.

[0089] Table 1 Node selection table for eight-neighborhood search method

[0090]

[0091] When using the sixteen-neighborhood search, as shown in Table 2, according to the range of the angle α, 9 invalid nodes in the sixteen-neighborhood are removed and 7 valid nodes are retained.

[0092] Table 2 Node selection table of the new sixteen-neighborhood search method

[0093]

[0094] In the embodiment of the present invention, the invalid nodes in the algorithm search process are removed through the node optimization strategy, thereby reducing the search time of the path planning algorithm.

[0095] Step S3, adopting a bidirectional search strategy and adding a dynamic weight coefficient adjustment heuristic function to obtain the planned path.

[0096] It is understandable that although the search speed of the traditional A* algorithm has been greatly improved compared to other heuristic algorithms, the search speed is still slow, which will significantly reduce the search efficiency when searching in a large-scale map environment. Therefore, the present invention adopts a two-way search strategy to reduce the number of nodes traversed by the algorithm and improve the search speed of the algorithm.

[0097] Bidirectional search is divided into forward search and reverse search. After determining the coordinates of the starting point and the end point, the forward search searches from the starting point to the end point, and the reverse search searches from the end point to the starting point. The current node of the algorithm is the parent node defined as Pnode, and the adjacent search node is the child node defined as Snode.

[0098] In step S3, the bidirectional search strategy may specifically include the following steps S31 to S34:

[0099] S31, initializing the forward search and reverse search lists; the forward search searches from the starting point to the end point, and the reverse search searches from the end point to the starting point.

[0100] Specifically, in the actual operation process, initialize Openlist1 and Closelist1 for forward search. Add the starting point to Closelist1 and define it as the current parent node Pnode1. Initialize Openlist2 and Closelist2 for reverse search. Add the end point to Closelist2 and define it as the current parent node Pnode2.

[0101] The Chinese definitions of relevant terms are as follows:

[0102] Openlist1: The open list for forward search, used to store nodes to be explored.

[0103] Closelist1: The closed list of the forward search, used to store the nodes that have been explored.

[0104] Openlist2: An open list for reverse search, used to store nodes to be explored (starting from the end point).

[0105] Closelist2: The closed list of reverse search, used to store the explored nodes (starting from the end point).

[0106] Pnode1: The current parent node of the forward search.

[0107] SPnode1: The neighboring nodes around Pnode1, that is, the child nodes to be evaluated in the forward search.

[0108] S32, calculating the cost value of each node in the list according to the cost calculation function, and selecting the node with the smallest cost value as the node to be explored next.

[0109] Specifically, the forward search process includes: defining the adjacent nodes around Pnode1 as SPnode1, and adding these nodes to Openlist1. Calculating the cost values ​​of all nodes in Openlist1 according to the cost calculation function. Selecting the node with the smallest cost value as the node to be explored next, defining it as SPnode2, and deleting it from Openlist1 and adding it to Closelist1.

[0110] The reverse search process includes: defining the neighboring nodes around Pnode2 as SRnode1 and adding these nodes to Openlist2. Calculating the cost value of all nodes in Openlist2 according to the cost calculation function. Selecting the node with the smallest cost value as the node to be explored next, defining it as SRnode2, deleting it from Openlist2 and adding it to Closelist2.

[0111] S33, determine whether the forward search and the reverse search intersect, if so, the algorithm ends, otherwise, continue searching.

[0112] Specifically, check whether SPnode2 exists in Openlist2 and whether SRnode2 exists in Closelist1. If so, it means that the forward search and the reverse search intersect at one point, and the algorithm ends at this time, and the optimal path is found. If not, it means that the forward search and the reverse search have not intersected, and it is necessary to redefine SPnode2 as the new parent node Pnode1 and SRnode2 as the new parent node Pnode2, and then repeat steps S31 and S32 until the forward and reverse searches intersect at one point.

[0113] S34, during the bidirectional search process, the heuristic function is adjusted using a dynamic weight coefficient to dynamically adjust the search speed.

[0114] It is understandable that in a two-way search, if the starting point and the end point are far apart, there may be a situation where the forward search and the reverse search intersect or fail to intersect, resulting in reduced efficiency and accuracy of the algorithm. To solve this problem, the present invention uses a dynamic weight coefficient w(n) to adjust the heuristic function h(n), dynamically balance the proportional relationship between g(n) and h(n), and thus dynamically control the search speed and search range of the path planning algorithm. Among them, g(n) represents the cost value (actual cost) from the starting point to the current node n, and h(n) represents the cost value (heuristic cost) from the current node n to the end point.

[0115] The calculation formula for the total cost of the bidirectional search algorithm after using the dynamic weight coefficient w(n) is:

[0116] f(n)=g(n)+w(n)·h(n) (4)

[0117] In the formula, f(n) represents the total cost from the starting point to the end point;

[0118] Among them, in the forward search, the expression of the dynamic weight coefficient is:

[0119]

[0120] During reverse search, the expression of the dynamic weight coefficient is:

[0121]

[0122] In the formula, x s and s is the starting point coordinate, x g and g is the end point coordinate, x t and t is the current coordinate of the forward search, x u and u The current coordinates for reverse search;

[0123] The dynamic weight coefficient w(n) adopts a combination of exponential and logarithmic functions to dynamically adjust the search speed according to the distance between the current node and the target endpoint; if the distance between the current node and the target endpoint is greater than the preset distance, the search speed is accelerated; if the distance between the current node and the target endpoint is less than the preset distance, the search speed is reduced to improve the search accuracy.

[0124] In the embodiment of the present invention, by introducing a two-way search mechanism, the efficiency of the algorithm search is improved, and at the same time, a dynamic weight factor is added to adjust the algorithm search speed, reduce the search of invalid areas, thereby improving the search efficiency and ensuring that the two-way search can meet at the middle point.

[0125] Step S4, adopting a redundant inflection point deletion strategy for the planned path, removing useless inflection points, and obtaining the optimal path.

[0126] After the A* algorithm plans the optimal path, there will still be many useless turning points. To address this problem, the present invention adopts a redundant turning point deletion strategy for the planned path, and effectively removes useless turning points by judging whether there are obstacles between nodes.

[0127] Step S4 specifically includes the following steps S41 to S44. Figure 6 The redundant node deletion schematic diagram provided by the present invention is combined with Figure 6 The redundant inflection point deletion strategy provided by the present invention is described as follows:

[0128] S41, connecting the start node forward to the inflection points separated therefrom, and determining whether there is an obstacle in the line connecting the start node and the inflection points separated therefrom; wherein the end point of the planned path is taken as the initial start node.

[0129] Figure 6 Where P1 is the starting point of the planned path, and P10 is the end point of the planned path. The end point P10 is used as the initial starting node.

[0130] S42, if there is no obstacle, delete the inflection point between the start node and the separated inflection point, and return to S41; if there is an obstacle, take the separated inflection point as an indelible inflection point, take the obtained indelible inflection point as a new start node, and return to S41.

[0131] Reference Figure 6 , starting from the end point P10, connect forward to the inflection point P8, and determine whether there is an obstacle in the connection line. If there is no obstacle, delete the inflection point P9 node between the start node P10 and the inflection point P8.

[0132] If there is an obstacle, the inflection point P8 is used as an inflection point that cannot be deleted, and the new inflection point P8 is used as a new starting node. Return to S41, connect the starting node P8 to the inflection point P6 that is separated from it, and determine whether there is an obstacle in the line connecting the starting node P8 and the inflection point P6 that is separated.

[0133] S43, repeating S41 to S42, and gradually processing forward to the starting point of the path;

[0134] S44, reconnecting all the obtained non-deletable inflection points to form an optimal path.

[0135] In this embodiment, useless turning points are removed by using a redundant turning point deletion strategy, thereby improving the driving efficiency of the mobile robot.

[0136] Figure 7The structure diagram of the mobile robot path planning system provided by the present invention is shown in FIG. Figure 7 , the mobile robot path planning system 700 includes:

[0137] Neighborhood search optimization module 701, used to determine a multi-neighborhood hybrid search mechanism using a mixture of eight neighborhoods and new sixteen neighborhoods according to whether there are obstacles in the eight neighborhoods of the current node;

[0138] A node optimization module 702 is used to delete invalid nodes according to the positions of the current node and the target node based on the multi-neighborhood hybrid search mechanism;

[0139] A bidirectional search module 703 is used to adopt a bidirectional search strategy and add a dynamic weight coefficient adjustment heuristic function to obtain a planned path;

[0140] The redundant inflection point deletion module 704 is used to adopt a redundant inflection point deletion strategy for the planned path, remove useless inflection points, and obtain the optimal path.

[0141] The mobile robot path planning system provided by the present invention executes the mobile robot path planning method provided by the above-mentioned embodiments through the above-mentioned modules. The mobile robot path planning method has been described in detail in the above-mentioned embodiments, and will not be repeated in this embodiment.

[0142] Figure 8 The structural block diagram of the electronic device provided by the present invention is as follows: Figure 8 As shown, the present invention also provides an electronic device, and the electronic device 800 can be a computing device such as a mobile terminal, a desktop computer, a notebook, a palm computer, and a server. The electronic device 800 includes a processor 801 and a memory 802, wherein the memory 802 stores a mobile robot path planning program 808.

[0143] In some embodiments, the memory 802 may be an internal storage unit of a computer device, such as a hard disk or memory of the computer device. In other embodiments, the memory 802 may also be an external storage device of the computer device, such as a plug-in hard disk equipped on the computer device, a smart memory card (Smart Media Card, SMC), a secure digital (SecureDigital, SD) card, a flash card (Flash Card), etc. Furthermore, the memory 802 may also include both an internal storage unit of the computer device and an external storage device. The memory 802 is used to store application software and various types of data installed on the computer device, such as program codes installed on the computer device. The memory 802 may also be used to temporarily store data that has been output or is to be output. In one embodiment, when the mobile robot path planning program 808 is executed by the processor 801, the following steps are implemented:

[0144] S1, according to whether there are obstacles in the eight neighborhoods of the current node, determine the multi-neighborhood hybrid search mechanism using the eight neighborhoods and the new sixteen neighborhoods;

[0145] S2, based on the multi-neighborhood hybrid search mechanism, delete invalid nodes according to the positions of the current node and the target node;

[0146] S3, adopts a bidirectional search strategy and adds a dynamic weight coefficient adjustment heuristic function to obtain the planned path;

[0147] S4, adopt the redundant inflection point deletion strategy for the planned path, remove useless inflection points and obtain the optimal path.

[0148] In some embodiments, the processor 801 may be a central processing unit (CPU), a microprocessor or other data processing chip, used to run program codes or process data stored in the memory 802, such as executing a mobile robot path planning program.

[0149] This embodiment further provides a computer-readable storage medium on which a mobile robot path planning program is stored. When the mobile robot path planning program is executed by a processor, the following steps are implemented:

[0150] S1, according to whether there are obstacles in the eight neighborhoods of the current node, determine the multi-neighborhood hybrid search mechanism using the eight neighborhoods and the new sixteen neighborhoods;

[0151] S2, based on the multi-neighborhood hybrid search mechanism, delete invalid nodes according to the positions of the current node and the target node;

[0152] S3, adopts a bidirectional search strategy and adds a dynamic weight coefficient adjustment heuristic function to obtain the planned path;

[0153] S4, adopt the redundant inflection point deletion strategy for the planned path, remove useless inflection points and obtain the optimal path.

[0154] The above-mentioned embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

[0155] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A mobile robot path planning method, characterized in that: include: S1, according to whether there are obstacles in the eight neighborhoods of the current node, determine the multi-neighborhood hybrid search mechanism using the eight neighborhoods and the new sixteen neighborhoods; S2, based on the multi-neighborhood hybrid search mechanism, delete invalid nodes according to the positions of the current node and the target node; S3, adopts a bidirectional search strategy and adds a dynamic weight coefficient adjustment heuristic function to obtain the planned path; S4, adopt the redundant inflection point deletion strategy for the planned path, remove useless inflection points and obtain the optimal path.

2. The mobile robot path planning method according to claim 1, characterized in that: Before step S1, the method further includes: The grid method is used to simulate the environment and obtain a grid map for mobile robot path planning.

3. The mobile robot path planning method according to claim 2, characterized in that: In step S1, according to whether there are obstacles in the eight neighborhoods of the current node, a multi-neighborhood hybrid search mechanism using the eight neighborhoods and the new sixteen neighborhoods is determined, including: Determine whether there are obstacles in the eight neighborhoods of the current node: If there are obstacles, the eight-neighborhood search method is used; If there are no obstacles, a sixteen-neighborhood search method is used; wherein the sixteen-neighborhood search method is optimized based on an eight-neighborhood search method and a twenty-four-neighborhood search method.

4. The mobile robot path planning method according to claim 3, characterized in that: In step S2, based on the multi-neighborhood hybrid search mechanism, invalid nodes are deleted according to the positions of the current node and the target node, including: Determine the positions of the current node and the target node in the grid map, and calculate the angle α between the line connecting the current node and the target node and the positive direction of the x-axis of the grid map; Invalid nodes are removed according to the range of angle α and the current neighborhood search method: when using the eight-neighborhood search method, three invalid nodes in the eight-neighborhood are removed according to the range of angle α, and five valid nodes are retained; when using the sixteen-neighborhood search method, nine invalid nodes in the sixteen-neighborhood are removed according to the range of angle α, and seven valid nodes are retained.

5. The mobile robot path planning method according to claim 4, characterized in that: In step S3, a bidirectional search strategy is adopted and a dynamic weight coefficient adjustment heuristic function is added to obtain the planned path, which specifically includes: Initialize the forward search and reverse search lists; forward search searches from the starting point to the end point, and reverse search searches from the end point to the starting point; Calculate the cost value of each node in the list according to the cost calculation function, and select the node with the smallest cost value as the node to be explored next; Determine whether the forward search and the reverse search intersect. If so, the algorithm ends, otherwise, continue searching; During the bidirectional search process, the dynamic weight coefficient is used to adjust the heuristic function to dynamically adjust the search speed.

6. The mobile robot path planning method according to claim 5, characterized in that: In the bidirectional search process, the dynamic weight coefficient is used to adjust the heuristic function to dynamically adjust the search speed, including: In the bidirectional search process, the dynamic weight coefficient w(n) is used to adjust the heuristic function h(n) to dynamically balance the proportional relationship between g(n) and h(n); where g(n) represents the cost from the starting point to the current node n, and h(n) represents the cost from the current node n to the end point; The calculation formula for the total cost of the bidirectional search algorithm after using the dynamic weight coefficient w(n) is: f(n)=g(n)+w(n)·h(n) In the formula, f(n) represents the total cost from the starting point to the end point; Among them, in the forward search, the expression of the dynamic weight coefficient is: During reverse search, the expression of the dynamic weight coefficient is: In the formula, x s and s is the starting point coordinate, x g and g is the end point coordinate, x t and t is the current coordinate of the forward search, x u and u The current coordinates for reverse search; The dynamic weight coefficient w(n) adopts a combination of exponential and logarithmic functions to dynamically adjust the search speed according to the distance between the current node and the target endpoint; if the distance between the current node and the target endpoint is greater than the preset distance, the search speed is accelerated; if the distance between the current node and the target endpoint is less than the preset distance, the search speed is reduced to improve the search accuracy.

7. The mobile robot path planning method according to claim 5, characterized in that: In step S4, the redundant inflection point deletion strategy is adopted for the planned path to remove useless inflection points, including: S41, connecting the start node forward to the inflection points separated therefrom, and determining whether there is an obstacle in the line connecting the start node and the inflection points separated therefrom; wherein the end point of the planned path is taken as the initial start node; S42, if there is no obstacle, delete the inflection point between the start node and the separated inflection point, and return to S41; if there is an obstacle, take the separated inflection point as an indelible inflection point, take the obtained indelible inflection point as a new start node, and return to S41; S43, repeating S41 to S42, and gradually processing forward to the starting point of the path; S44, reconnecting all the obtained indelible turning points to form an optimal path.

8. A mobile robot path planning system, characterized in that: include: A neighborhood search optimization module is used to determine a multi-neighborhood hybrid search mechanism using a mixture of eight neighborhoods and new sixteen neighborhoods according to whether there are obstacles in the eight neighborhoods of the current node; A node optimization module, used to delete invalid nodes according to the positions of the current node and the target node based on the multi-neighborhood hybrid search mechanism; A bidirectional search module is used to adopt a bidirectional search strategy and add a dynamic weight coefficient to adjust the heuristic function to obtain the planned path; The redundant inflection point deletion module is used to adopt a redundant inflection point deletion strategy for the planned path, remove useless inflection points, and obtain the optimal path.

9. An electronic device, It is characterized in that comprising a memory and a processor, wherein: The memory is used to store programs; The processor is coupled to the memory and is used to execute the program stored in the memory to implement the steps in the mobile robot path planning method described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: Used to store computer-readable programs or instructions, which, when executed by a processor, can implement the steps in the mobile robot path planning method described in any one of claims 1 to 7.

Images