Mobile robot path planning method, device, equipment and medium

Abstract

The invention provides a mobile robot path planning method, device, equipment and medium, and relates to the technical field of path planning, and the method comprises the steps: guiding the expansion direction of a random tree of an RRT * algorithm through an artificial potential field method according to a starting point and a target point of a mobile robot, preferentially selecting sampling points of the random tree, and deleting redundant nodes, generating a global planning path; determining a current position and a local target point based on the global planning path; and according to the current position and the local target point, adopting the artificial potential field method and the simulated annealing algorithm strategy to generate a local planning path. The method can solve the problems that an RRT * algorithm is poor in guidance and difficult to approach to a target point quickly, rejects invalid nodes far away from a target, improves the sampling pertinence, and can guide a robot to be separated from a local extreme value.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of path planning, in particular to a mobile robot path planning method, device, equipment and medium. BACKGROUND

[0002] The mobile robot refers to an intelligent device capable of autonomously detecting surrounding conditions in a space environment, freely shuttling and completing various specified tasks. The path planning is a key component of the autonomous navigation of the mobile robot, which refers to a process of finding a best or feasible path from a starting position to a target position according to certain performance indicators (such as the shortest path, the least time, etc.) in a given environment.

[0003] In the related art, a heuristic algorithm such as an RRT algorithm, an RRT* algorithm and an artificial potential field method is generally used for path planning. However, the existing heuristic algorithm has problems such as poor guidance of random sampling, non-ideal length of generated path, redundant path nodes and final path in a polyline shape, and cannot obtain an optimal path planning scheme. SUMMARY

[0004] The problem solved by the present application is how to obtain an optimal path planning scheme.

[0005] To solve the above problems, the present application provides a mobile robot path planning method, device, equipment and medium.

[0006] In a first aspect, the present application provides a mobile robot path planning method, comprising: S1, according to the starting point and the target point of the mobile robot, guiding the expansion direction of the random tree of the RRT* algorithm by the artificial potential field method, optimally selecting the sampling point of the random tree, deleting the redundant nodes, and generating a global planning path; S2, determining a current position and a local target point based on the global planning path; S3, according to the current position and the local target point, generating a local planning path by using the artificial potential field method and the simulated annealing algorithm strategy.

[0007] Optionally, S1, according to the starting point and the target point of the mobile robot, guiding the expansion direction of the random tree of the RRT* algorithm by the artificial potential field method, optimally selecting the sampling point of the random tree, deleting the redundant nodes, and generating a global planning path, comprises: S11, initializing the random tree, setting basic parameters of the random number according to the starting point and the target point; S12, determining a guided random sampling point and a neighboring node corresponding to the guided random sampling point by guiding an expansion direction of the random tree through the artificial potential field method, and determining a new sampling point according to the guided random sampling point and the neighboring node; S13, preferentially selecting the new sampling point to obtain a reserved new sampling point; S14, performing collision detection on the reserved new sampling point and the neighboring node, when collision occurs, performing step S12, when no collision occurs, taking the reserved new sampling point as a new sampling point without collision, and adding the new sampling point without collision to the random tree; S15, sequentially performing parent node reselection, rewiring and collision detection on the new sampling point without collision to obtain a new random tree structure; S16, repeating steps S12-S15 until the new random tree structure reaches the target point within the effective range; S17, connecting the new sampling point without collision with the target point to obtain an initial path, deleting the redundant nodes and corresponding branches on the initial path to obtain the global planning path.

[0008] Optionally, the guiding of the expansion direction of the random tree of the RRT* algorithm through the artificial potential field method comprises: According to the guided random sampling point formula constructed by the artificial potential field method, the expansion direction of the random tree is guided, and the guided random sampling point formula comprises: ; Wherein, Q new is the guided random sampling point, Q rand is a random sampling point, Q near is a neighboring node of the random tree, Q goal is the target point, L (n) is a first attractive force function, G (n) is a second attractive force function, m is a first attractive factor threshold, n is a second attractive factor threshold, and the sum of m and n is 1, α is an expansion step length.

[0009] Optionally, the S13, preferentially selecting the sampling point to obtain a reserved new sampling point, comprises: S131, respectively determining a first distance from the neighboring node to the target point and a second distance from the new sampling point to the target point; S132, when the first distance is greater than the second distance, taking the new sampling point as the reserved sampling point, when the first distance is less than or equal to the second distance, rejecting the new sampling point, and returning to step S12.

[0010] Optionally, the deleting the redundant nodes and corresponding branches on the initial path comprises: determining the Euclidean distances of each node on the initial path to the target point and comparing to obtain the maximum Euclidean distance of the initial path; determining the redundant Euclidean distances of other nodes on the random tree to the target point, comparing each of the redundant Euclidean distances with the maximum Euclidean distance, and deleting the nodes and corresponding branches thereof whose redundant Euclidean distances are greater than the maximum Euclidean distance.

[0011] Optionally, the S3 comprises: generating a local planning path according to the current position and the local target point by using the artificial potential field method and the simulated annealing algorithm strategy. S31, determining a distance value of the current position and the local target point, and determining the force size of the current position by using the artificial potential field method according to the distance value; S32, when the force size is zero, performing step S33, and when the force size is not zero, performing step S34; S33, generating a temporary target point by using the simulated annealing algorithm, taking the temporary target point as a new local target point, and returning to step S31; S34, determining a next path point by using the artificial potential field method according to the force size, taking the next path point as a new current position, returning to step S1, and when the next path point is the same as the local target point, generating the local planning path according to each of the next path points.

[0012] Optionally, the determining the force size of the current position by using the artificial potential field method according to the distance value comprises: when the distance value is greater than a preset distance threshold, determining the force size by using a distance influence factor and a distance power term; when the distance value is less than the preset distance threshold, determining the force size by using the artificial potential field method.

[0013] In a second aspect, the present application provides a mobile robot path planning device, comprising: a global module configured to guide the expansion direction of a random tree of an RRT* algorithm by using an artificial potential field method according to the starting point and the target point of a mobile robot, to select a sampling point of the random tree, to delete a redundant node, and to generate a global planning path; a position module configured to determine a current position and a local target point based on the global planning path; A local module is configured to generate a local planning path by using the artificial potential field method and the simulated annealing algorithm strategy according to the current position and the local target point.

[0014] In a third aspect, the present application provides an electronic device comprising a memory and a processor. The memory is configured to store a computer program. The processor is configured to implement the mobile robot path planning method according to the first aspect when executing the computer program.

[0015] In a fourth aspect, the present application provides a computer readable storage medium, wherein the storage medium stores a computer program, and the computer program, when executed by a processor, implements the mobile robot path planning method according to the first aspect.

[0016] The mobile robot path planning method, device, equipment and medium provided by the present application have the following advantages: According to the starting point and the target point of the mobile robot, the artificial potential field method is used to guide the extension direction of the random tree of the RRT* algorithm, which can solve the problem that the RRT* algorithm is difficult to quickly approach the target point due to poor guidance. By selecting the sampling point of the random tree and deleting redundant nodes, the invalid nodes far from the target can be removed, the sampling pertinence can be improved, the redundant nodes and their branches with a distance greater than a threshold value can be deleted, and the invalid search can be reduced, so as to improve the efficiency and accuracy of path planning, obtain the optimal global planning path, and use the artificial potential field method and the simulated annealing algorithm strategy according to the current position and the local target point. When the mobile robot is trapped in a local extremum region, for example, when the attraction and repulsion are balanced to cause stagnation, a temporary virtual target point is generated near the current position by the simulated annealing algorithm, the force field balance is broken, and the robot is guided to escape from the local extremum, so as to complete local path planning and improve the efficiency of path planning, thereby quickly obtaining a local planning path. BRIEF DESCRIPTION OF DRAWINGS

[0017] Figure 1 A flowchart of the mobile robot path planning method provided by the embodiment of the present application is shown in the figure. Figure 2 A schematic diagram of the global planning path provided by the embodiment of the present application is shown in the figure. Figure 3 A schematic diagram of the guided random sampling point provided by the embodiment of the present application is shown in the figure. Figure 4 A structural schematic diagram of the mobile robot path planning device provided by the embodiment of the present application is shown in the figure. Figure 5 A structural schematic diagram of the electronic device provided by the embodiment of the present application is shown in the figure. DETAILED DESCRIPTION

[0018] In order to make the above objectives, characteristics and advantages of the present application more obvious and comprehensible, specific embodiments of the present application are described in detail below with reference to the drawings. Although some embodiments of the present application are shown in the drawings, it should be understood that the present application can be implemented in various forms, and should not be interpreted as being limited to the embodiments described herein, but rather, these embodiments are provided to make the present application more thorough and complete. It should be understood that the drawings and embodiments of the present application are only for illustrative purposes, and are not intended to limit the scope of protection of the present application.

[0019] It should be understood that each step described in the method embodiments of the present application can be performed in different orders and / or in parallel. In addition, the method embodiments can include additional steps and / or omit the steps shown. The scope of the present application is not limited in this respect.

[0020] The term "comprising" and variations thereof as used herein are open-ended, that is "including, but not limited to"; the term "based on" is "based, at least in part, on"; the term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments"; the term "optional" means "optional in at least some embodiments". Related definitions are given throughout the description. It should be noted that the concepts "first", "second", etc. mentioned in the present application are only used to distinguish different devices, modules or units, and are not intended to limit the order or interdependence of the functions performed by these devices, modules or units.

[0021] It should be noted that the modification of "one" or "multiple" mentioned in the present application is illustrative rather than limiting, and those skilled in the art should understand that, unless otherwise explicitly indicated in the context, it should be understood as "one or more".

[0022] The names of the messages or information exchanged between the devices in the embodiments of the present application are only for illustrative purposes, and are not intended to limit the scope of the messages or information.

[0023] To solve the problems of the related art described above, the present embodiment provides a mobile robot path planning method, device, equipment and medium.

[0024] As shown in Figure 1 The mobile robot path planning method provided by the present embodiment includes: S1, according to the starting point and target point of the mobile robot, guiding the extension direction of the random tree of the RRT* algorithm by the artificial potential field method, optimally selecting the sampling point of the random tree, and deleting the redundant nodes to generate a global planning path.

[0025] Specifically, before path planning, environment modeling needs to be performed. The mobile robot collects data such as geometric characteristics, spatial positions, and appearance attributes of objects in the physical environment via various sensing devices, visual sensors, laser range finders, ultrasonic detection devices, and the like, and then converts the environment data into a digital environment model to support the environment perception and motion planning of the robot. Common environment modeling methods such as topological graph method, visual graph method, and grid method can be used for environment modeling. The artificial potential field method is a path planning method based on physical field simulation. By constructing the "gravitational force" of the target point on the robot and the "repulsive force" of the obstacle on the robot, the path planning problem is converted into a particle motion problem in the combined force field. RRT* is a global path planning algorithm based on random sampling. The feasible path is searched by iteratively expanding the random tree structure. The RRT* algorithm in the prior art has strong randomness in the sampling process. Therefore, during the path planning process of searching the target point, the search range is too large, which leads to low search efficiency and a large amount of search time. Therefore, in order to improve the search efficiency, the target point is used to select the optimal random sampling point, and the redundant points are deleted based on the RRT* algorithm in the prior art. In addition, although the RRT* algorithm after the optimal selection and deletion of the redundant points reduces the calculation amount, due to the generally low target-oriented probability setting, a large number of invalid expansion nodes are still generated. To solve this contradiction, the artificial potential field method is introduced in this embodiment. The expansion direction of the random tree is guided by establishing the target point gravitational field, which not only retains the probability completeness advantage of the RRT* algorithm, but also enhances the target-oriented expansion by the gravitational effect. Not only can the number of invalid nodes generated be significantly reduced, but also the overall planning time can be effectively shortened. The global planning path is shown in FIG. 1, wherein the lower left corner is the starting point, the upper right corner is the target point, and the black area is the obstacle. Figure 2

[0026] S2, determining a current position and a local target point based on the global planning path.

[0027] Specifically, after determining the global planning path, the mobile robot moves based on the global planning path. During the movement, local path planning needs to be performed according to the global planning path, taking the current position of the mobile robot as the starting point of the local path planning and taking the path point in the global planning path as the target point of the local path planning.

[0028] S3, generating a local planning path by using the artificial potential field method and the simulated annealing algorithm strategy according to the current position and the local target point.

[0029] ​Specifically, according to the current position and local target point, the artificial potential field method is used for local path planning. When the artificial potential field method falls into a local minimum, the simulated annealing algorithm strategy is adopted to randomly set virtual target points in the local path planning process, thereby guiding the mobile robot to jump out of the local potential field trap, improving the global exploration ability of the artificial potential field path planning of the existing technology, providing a local path planning effect, and obtaining a local planned path.

[0030] In this embodiment, according to the starting point and target point of the mobile robot, the expansion direction of the random tree of the RRT* algorithm is guided by the artificial potential field method, which can solve the problem that the RRT* algorithm has poor guidance and is difficult to quickly approach the target point. By preferentially selecting the sampling points of the random tree and deleting redundant nodes, invalid nodes far from the target can be eliminated, thereby improving the sampling targeting. Redundant nodes and their branches with a distance greater than a threshold are deleted to reduce invalid searches, thereby improving the efficiency and accuracy of path planning and obtaining the optimal global planning path. According to the current position and the local target point, the artificial potential field method and simulated annealing algorithm strategy are adopted. When the mobile robot falls into a local extreme value area, for example, when the balance of gravity and repulsion causes stagnation, a temporary virtual target point is generated near the current position through the simulated annealing algorithm to break the force field balance and guide the robot out of the local extreme value, thereby completing local path planning and improving the efficiency of path planning, thereby quickly obtaining the local planning path.

[0031] Optionally, the step S1, based on the starting point and target point of the mobile robot, guides the expansion direction of the random tree of the RRT* algorithm by an artificial potential field method, preferentially selects sampling points of the random tree, deletes redundant nodes, and generates a global planning path, including: S11, initializing the random tree, and setting basic parameters of the random number according to the starting point and the target point; S12, guiding the expansion direction of the random tree by the artificial potential field method, determining a directional random sampling point and an adjacent node corresponding to the directional random sampling point, and determining a new sampling point based on the directional random sampling point and the adjacent node; S13, selecting the new sampling point based on merit to obtain the reserved new sampling point; S14, performing collision detection on the retained new sampling point and the adjacent node. When a collision occurs, executing step S12; when no collision occurs, taking the retained new sampling point as a new sampling point without collision and adding it to the random tree; S15, performing parent node reselection, rewiring, and collision detection on the collision-free new sampling points in sequence to obtain a new random tree structure; S16, looping steps S12-S15 until the collision-free sampling points of the new random tree structure reach the range of the target point; S17, connecting the collision-free sampling point and the target point to obtain an initial path, deleting the redundant nodes and corresponding branches on the initial path to obtain the global planning path.

[0032] Specifically, first, a random tree is initialized in the constructed environment model, the start point and the target point of the RRT* algorithm are defined according to the start point and the target point, the root start node of the random tree is selected as the start point, and the basic parameters of the random number are set, for example, the step length and the iteration number; the expansion direction of the random tree is guided by the artificial potential field method, the directional random sampling point is determined, all tree nodes in the random tree are traversed, the Euclidean distance between the directional random sampling point and each tree node is calculated, the nearest node corresponding to the directional random sampling point is taken as the node with the minimum Euclidean distance, the directional random sampling point and the nearest node are connected, and a new sampling point is determined in the connecting line from the nearest node in the direction away from the directional random sampling point with the step length as the cost; the new sampling point is selected optimally to obtain a reserved new sampling point; the reserved new sampling point and the nearest node are subjected to collision detection by using a collision detection function, whether the connecting line of the new sampling point and the nearest node collides with the obstacle is judged, when collision occurs, step S12 is executed, when no collision occurs, the reserved new sampling point is taken as a collision-free new sampling point and added to the random tree; the collision-free new sampling point is sequentially subjected to parent node reselection, rewiring and collision detection, the parent node reselection step includes: drawing a circle with the collision-free new sampling point as the center and a preset value as the radius, traversing all nodes in the circle, calculating the path cost of the start point passing through all nodes in the circle to the collision-free new sampling point, taking the node with the minimum cost as the parent node of the collision-free new sampling point, and obtaining a new random tree structure after rewiring and collision detection; the steps S12-S15 are looped until the collision-free sampling point of the new random tree structure reaches the action range of the target point, then the collision-free sampling point and the target point are connected to obtain an initial path, the redundant nodes and corresponding branches on the initial path are deleted to obtain the global planning path.

[0033] Illustratively, when the iteration number of the random tree expansion exceeds the preset value and the condition for stopping sampling is still not met, it is determined that the current path planning is unsuccessful, otherwise, it returns to step S15. Path backtracking is performed, the target point is taken as the starting point, and the parent node pointer is traced back to the start point, that is, a feasible path connecting the start point and the target point is obtained as the global planning path.

[0034] Optionally, the expansion direction of the random tree of the RRT* algorithm guided by the artificial potential field method comprises: The expansion direction of the random tree is guided according to the directional random sampling point formula constructed by the artificial potential field method, and the directional random sampling point formula comprises: ; wherein Q new is the guided random sampling point, Q rand is the random sampling point, Q near is the neighboring node of the random tree, Q goal is the target point, L (n) is the first attractive function, m is a first attractive factor threshold, and n is a second attractive factor threshold, G (n) is the second attractive function, m is a first attractive factor threshold, and n is a second attractive factor threshold, and the sum of m and n is 1, α is the step length.

[0035] Specifically, m L (n) + n G (n) is the growth guide function of the guided random sampling point, by adjusting m and n, the target bias of the neighboring node and the target point to the guided random sampling point can be changed. G (n) is the second attractive function, which is also the target attractive function of the RRT* algorithm. When the guided random sampling point is expanded, the second attractive function affects the value of the guided random sampling point through the attractive force of the neighboring node and the target point, thereby guiding the growth of the random tree to the target point. After the artificial potential field method is used to guide the expansion direction of the random tree of the RRT* algorithm, the random tree sampling mode is no longer random sampling of the whole graph, nor is it determined according to the random point position and the step length alone, but is determined by the addition of the random point with different weights and the target point attractive force vector. At this time, the generated guided random sampling point is as shown in Figure 3 .

[0036] Optionally, the S13, the sampling point is selected, and a reserved new sampling point is obtained, including: S131, respectively determining a first distance from the neighboring node to the target point and a second distance from the new sampling point to the target point; S132, when the first distance is greater than the second distance, the new sampling point is taken as the reserved sampling point, and when the first distance is less than or equal to the second distance, the new sampling point is removed, and the step S12 is returned.

[0037] Specifically, the Euclidean distance from the neighboring node to the target point, i.e. the first distance, and the Euclidean distance from the new sampling point to the target point, i.e. the second distance, are determined respectively. When the first distance is greater than the second distance, the new sampling point is taken as the reserved sampling point, and when the first distance is less than or equal to the second distance, the new sampling point is removed, and the step S12 is returned. The redundant points of the RRT* algorithm are removed, thereby reducing the complexity and running time of the algorithm and increasing the efficiency.

[0038] Optionally, the deleting the redundant nodes and corresponding branches on the initial path comprises: determining the Euclidean distance of each node on the initial path to the target point and comparing to obtain the maximum Euclidean distance of the initial path; determining the redundant Euclidean distance of other nodes on the random tree to the target point, comparing each redundant Euclidean distance with the maximum Euclidean distance, and deleting the nodes and their corresponding branches whose redundant Euclidean distance is greater than the maximum Euclidean distance.

[0039] Specifically, the RRT* algorithm generates a large number of redundant nodes in the process of approaching the target point, therefore, it is necessary to prune the RRT* algorithm to reduce unnecessary search work and improve work efficiency. First, all nodes on the initial path are traversed, the Euclidean distance of each node on the initial path to the target point is determined by the maximum Euclidean distance formula and compared to obtain the maximum Euclidean distance of the initial path, and then the redundant Euclidean distance of other nodes on the random tree to the target point is determined by the redundant Euclidean distance formula, each redundant Euclidean distance is compared with the maximum Euclidean distance, and the nodes and their corresponding branches whose redundant Euclidean distance is greater than the maximum Euclidean distance are deleted. ; wherein D max is the maximum Euclidean distance, i is the node index on the initial path, t is the mapping point of the Q i node on the initial path on the line connecting the start and end points to the target point Q goal , and Q init is the start point.

[0040] The redundant Euclidean distance formula includes: ; wherein D Tj is the redundant Euclidean distance, and j is the index of the remaining nodes on the random tree.

[0041] Optionally, the S3 generates a local planning path according to the current position and the local target point by using the artificial potential field method and the simulated annealing algorithm strategy, comprising: S31, determining the distance value of the current position and the local target point, and determining the force size of the current position by using the artificial potential field method according to the distance value; S32, when the force size is zero, executing step S33, and when the force size is not zero, executing step S34; S33, generating a temporary target point by using the simulated annealing algorithm, taking the temporary target point as a new local target point, and returning to step S31; S34, according to the force size, the artificial potential field method is used to determine the next path point, and the next path point is taken as a new current position, returning to step S1, and when the next path point is the same as the local target point, the local planning path is generated according to each next path point.

[0042] Specifically, first, parameters such as a local starting point, a local target point, a current position of the robot, and obstacles are initialized. Then, a distance value of the current position and the local target point is determined, and the force size of the current position is determined by using the artificial potential field method according to the distance value; whether the robot is trapped in a local extreme point is judged according to the force size, and when the force size is zero, it is indicated that the robot is trapped in the local extreme point, and step S33 is executed, and when the force size is not zero, step S34 is executed; when the robot is trapped in the local extreme point, a temporary target point is generated by using a simulated annealing algorithm, the temporary target point is taken as a new local target point, and step S31 is returned to. The simulated annealing algorithm is a method for finding the best solution by using a random strategy, and when the mobile robot is trapped in an extreme value problem, the local extreme value trap is escaped by a method of randomly adding a target point, that is, the added target point can change the force size, so that the extreme value problem can be escaped, and obstacles can be effectively avoided. When the local extreme value trap is escaped or not trapped, the next path point is determined by using the artificial potential field method according to the force size, and the next path point is taken as a new current position, returning to step S1, and the cycle is performed, so that the local path planning is promoted, and when the next path point is the same as the local target point, that is, when the local target point is reached, the local planning path is generated according to each next path point. Exemplarily, the temporary target point is generated by using the simulated annealing algorithm, including: first, a random point is selected in a local minimum point region where the mobile robot is currently located, and then the potential fields of the current position of the mobile robot and the random point are calculated respectively, and the difference value of the two potential fields is determined, and if the difference value is less than zero, it is indicated that the random point is accepted and can be taken as a temporary target point; and if the difference value is greater than zero, whether the random point is accepted is determined according to a simulated annealing criterion.

[0043] Optionally, the force size of the current position is determined by using the artificial potential field method according to the distance value, including: When the distance value is greater than a preset distance threshold, the force size is determined by a distance influence factor and a distance power term. When the distance value is less than the preset distance threshold, the force size is determined by using the artificial potential field method.

[0044] Specifically, when the distance value is less than the preset distance threshold, the artificial potential field method is used to determine the size of the resultant force. When the distance value is greater than the preset distance threshold, the improved repulsive force function and the improved attractive force formula are used to determine the size of the resultant force, wherein the improved repulsive force function is obtained by improving the repulsive force function of the prior art artificial potential field method through the distance power term of the current position and the local target point, and the improved attractive force formula is obtained by adding a distance influence factor to the attractive force formula of the prior art artificial potential field method, so as to solve the collision problem caused by the excessive attractive force of the local target point when the mobile robot is far away from the local target point. The improved attractive force formula includes: ; wherein, is the size of the attractive force, the attractive force is the attractive force of the attractive potential field function , the derivative of the attractive potential field with respect to the relative distance between the target point and the mobile robot is the attractive force function, and the attractive force function is a linear function of the first order, is the attractive force gain coefficient, Z is the current position, Z g is the local target point, and d is the preset distance threshold, i.e. the distance influence factor.

[0045] The improved repulsive force function includes: ; wherein, is the size of the repulsive force, is the repulsive potential field function, and the repulsive potential field function includes: ; and are the first repulsive force and the second repulsive force respectively, and and are respectively: ; ; wherein, is the relative distance between the current position and the local target point, is the relative distance between the current position and the obstacle Z o , K rep is the repulsive force scale factor, is the influence range of the obstacle repulsive potential field, and n is any number greater than zero, a 1 and a 2 are respectively: ; .

[0046] It is worth mentioning that the repulsive potential field function is in accordance with the repulsive function of the traditional artificial potential field method, and is specially adjusted for the case that the target is unreachable. In the repulsive function, the n-th power term of the distance between the mobile robot and the local target point is used. When the mobile robot approaches the local target point, the growth rate of the repulsive force will be greatly slowed down, effectively solving the problem of excessive repulsive force, so that the robot can successfully reach the local target point under the joint action of the repulsive force and the attractive force. The repulsive function dynamically adjusts the strength of the repulsive field by means of the power term of the distance. When the mobile robot is far away from the local target point, the repulsive force remains a certain strength, so that the mobile robot can effectively avoid obstacles. When the mobile robot approaches the local target point, the growth rate of the repulsive force slows down, preventing the repulsive force from being too large to affect the mobile robot to reach the local target point. This improvement not only solves the problem that the target cannot be reached, but also improves the stability and adaptability of the algorithm. This adjustment not only retains the timeliness and intuitiveness of the traditional algorithm, but also greatly improves the suitability and reliability of the algorithm in complex environments.

[0047] As shown in Figure 4 The embodiment of the application provides a mobile robot path planning device, which comprises: A global module is configured to guide the extension direction of a random tree of an RRT* algorithm by an artificial potential field method according to a starting point and a target point of a mobile robot, to select a sampling point of the random tree, to delete redundant nodes, and to generate a global planning path. A position module is configured to determine a current position and a local target point based on the global planning path. A local module is configured to generate a local planning path by using the artificial potential field method and a simulated annealing algorithm strategy according to the current position and the local target point.

[0048] As shown in Figure 5 The embodiment of the application provides an electronic device 500, which comprises a memory 510 and a processor 520; the memory 510 is used for storing a computer program; and the processor 520 is used for realizing the mobile robot path planning method as described above when the computer program is executed.

[0049] Alternatively, an electronic device 500 comprises a memory 510 and a processor 520 coupled to the memory 510; the memory 510 is configured to store a computer program; and the processor 520 is configured to perform the following operations when the computer program is executed: S1, guiding the extension direction of a random tree of an RRT* algorithm by an artificial potential field method according to a starting point and a target point of a mobile robot, selecting a sampling point of the random tree, deleting redundant nodes, and generating a global planning path; S2, determining a current position and a local target point based on the global planning path. S3, generating a local planning path according to the current position and the local target point by using the artificial potential field method and a simulated annealing algorithm strategy.

[0050] The embodiment of the present application provides a computer readable storage medium, and the storage medium stores a computer program.

[0051] Alternatively, a non-volatile computer readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor executes the following operations: S1, generating a global planning path by guiding an expansion direction of a random tree of an RRT* algorithm by an artificial potential field method, optimally selecting a sampling point of the random tree, and deleting a redundant node according to a starting point and a target point of a mobile robot; S2, determining a current position and a local target point based on the global planning path; S3, generating a local planning path according to the current position and the local target point by using the artificial potential field method and a simulated annealing algorithm strategy.

[0052] An electronic device 500 that can be a server or a client of the present application will now be described, which is an example of a hardware device that can be applied to aspects of the present application. The electronic device 500 is intended to represent various forms of digital electronic computer devices such as laptops, desktops, tablets, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device 500 can also represent various forms of mobile devices such as personal digital processors, cellular telephones, smart phones, wearable devices, and other like computing devices. The components shown here, their connections and relationships, and their functions, are meant to be examples only, and are not meant to limit implementations of the present application described and / or claimed in this document.

[0053] The electronic device 500 includes a computing unit that can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) or a computer program loaded into a random access memory (RAM) from a storage unit. In the RAM, various programs and data required for device operation can also be stored. The computing unit, the ROM, and the RAM are connected to each other through a bus. An input / output (I / O) interface is also connected to the bus.

[0054] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by a computer program instructing relevant hardware, and the program can be stored in a computer readable storage medium. When the program is executed, the program can include the processes of the above-mentioned embodiment methods. The storage medium can be a magnetic disc, an optical disc, a read-only memory (ROM), a random access memory (RAM), or the like. In this application, the units described as separate components can or can not be physically separated, and the components shown as units can or can not be physical units, that is, they can be located in one place, or they can be distributed on multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiment of the present application. In addition, the functional units in each embodiment of the present application can be integrated in one processing unit, or each unit can exist physically independently, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0055] Although the present application is disclosed as above, the protection scope of the present application is not limited to this. Those skilled in the art can make various changes and modifications without departing from the spirit and scope of the present application, and these changes and modifications will fall within the protection scope of the present application.

Claims

1. A mobile robot path planning method, characterized in that: include: S1, according to the starting point and target point of the mobile robot, guide the expansion direction of the random tree of the RRT* algorithm through the artificial potential field method, select the sampling points of the random tree preferentially, delete redundant nodes, and generate a global planning path; S2, determining the current position and the local target point based on the global planning path; S3, generating a local planning path according to the current position and the local target point by using the artificial potential field method and simulated annealing algorithm strategy.

2. The mobile robot path planning method according to claim 1, characterized in that: Said S1, based on the starting point and target point of the mobile robot, guides the expansion direction of the random tree of the RRT* algorithm through the artificial potential field method, preferentially selects the sampling points of the random tree, deletes redundant nodes, and generates a global planning path, including: S11, initializing the random tree, and setting basic parameters of the random number according to the starting point and the target point; S12, guiding the expansion direction of the random tree by the artificial potential field method, determining a directional random sampling point and an adjacent node corresponding to the directional random sampling point, and determining a new sampling point based on the directional random sampling point and the adjacent node; S13, selecting the new sampling point based on merit to obtain the reserved new sampling point; S14, performing collision detection on the retained new sampling point and the adjacent node. When a collision occurs, executing step S12; when no collision occurs, taking the retained new sampling point as a new sampling point without collision and adding it to the random tree; S15, performing parent node reselection, rewiring, and collision detection on the collision-free new sampling points in sequence to obtain a new random tree structure; S16, looping steps S12-S15 until the collision-free sampling points of the new random tree structure reach the range of the target point; S17, connecting the collision-free sampling points and the target point to obtain an initial path, deleting the redundant nodes and corresponding branches on the initial path to obtain the global planning path.

3. The mobile robot path planning method according to claim 2, characterized in that: The step of guiding the expansion direction of the random tree by the artificial potential field method includes: The expansion direction of the random tree is guided by a directional random sampling point formula constructed according to the artificial potential field method. The directional random sampling point formula includes: ； Among them, Q new is the directional random sampling point, Q rand is a random sampling point, Q near is the adjacent node of the random tree, Q goal is the target point, L (n) is the first gravitational action function, G (n) is the second gravitational action function, m is the first gravitational factor threshold, n is the second gravitational factor threshold, and the sum of m and n is 1, α To expand the step length.

4. The mobile robot path planning method according to claim 2, characterized in that: The step S13, selecting the sampling points based on merit to obtain new sampling points to be retained, includes: S131, respectively determining a first distance from the adjacent node to the target point, and a second distance from the new sampling point to the target point; S132: When the first distance is greater than the second distance, the new sampling point is used as the reserved sampling point; when the first distance is less than or equal to the second distance, the new sampling point is removed, and the process returns to step S12.

5. The mobile robot path planning method according to claim 2, characterized in that: The deleting the redundant nodes and corresponding branches on the initial path includes: Determine the Euclidean distance from each node on the initial path to the target point and compare them to obtain the maximum Euclidean distance of the initial path; Determine the redundant Euclidean distances from other nodes on the random tree to the target point, compare each redundant Euclidean distance with the maximum Euclidean distance, and delete the nodes and their corresponding branches whose redundant Euclidean distances are greater than the maximum Euclidean distance.

6. The mobile robot path planning method according to claim 1, characterized in that: The step S3, generating a local planning path based on the current position and the local target point using the artificial potential field method and simulated annealing algorithm strategy, includes: S31, determining a distance value between the current position and the local target point, and determining a magnitude of a resultant force at the current position using the artificial potential field method based on the distance value; S32, when the magnitude of the resultant force is zero, execute step S33, and when the magnitude of the resultant force is not zero, execute step S34; S33, using the simulated annealing algorithm to generate a temporary target point, using the temporary target point as a new local target point, and returning to step S31; S34, according to the size of the resultant force, the artificial potential field method is used to determine the next path point, and the next path point is used as the new current position, returning to step S1, and when the next path point is the same as the local target point, the local planning path is generated according to each next path point.

7. The mobile robot path planning method according to claim 6, characterized in that: Determining the magnitude of the resultant force at the current position using the artificial potential field method based on the distance value includes: When the distance value is greater than a preset distance threshold, the magnitude of the resultant force is determined by the distance influence factor and the distance power term; When the distance value is less than a preset distance threshold, the artificial potential field method is used to determine the magnitude of the resultant force.

8. A mobile robot path planning device, characterized in that: include: The global module is used to guide the expansion direction of the random tree of the RRT* algorithm through the artificial potential field method according to the starting point and target point of the mobile robot, select the sampling points of the random tree preferentially, delete redundant nodes, and generate a global planning path; A location module, configured to determine a current location and a local target point based on the global planning path; The local module is used to generate a local planning path according to the current position and the local target point by adopting the artificial potential field method and simulated annealing algorithm strategy.

9. An electronic device, characterized in that: including memory and processor; The memory is used to store computer programs; The processor is configured to implement the mobile robot path planning method according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that The storage medium stores a computer program, and when the computer program is executed by the processor, the mobile robot path planning method according to any one of claims 1 to 7 is implemented.

Images