Path determination method and motion control method
By simplifying the confined space and obstacles into grid systems and optimizing the path, the problem of randomness and computational volume of path planning in three-dimensional space is solved, and efficient and accurate path planning of the robot in the confined space is achieved.
Patent Information
- Application Number
- CN202410741840.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-06-11
- Publication Date
- 2024-11-15
- Estimated Expiration
- 2044-06-11
AI Technical Summary
In three-dimensional space, it is difficult for the prior art to effectively plan the path of the robot in the confined space, resulting in a large randomness in path generation, long search time, large calculation amount, and difficult to find feasible paths that meet constraints.
Simplify confined space and obstacle objects into a raster system, determine the initial raster and target raster of the robot in the raster system, generate multiple initial paths based on constraints, and obtain the optimal target path through path optimization.
It improves the accuracy and efficiency of path planning, reduces the time consumption of path planning, and realizes the robot's efficient, timely and accurate reaching the target position in a confined space.
Smart Images

Figure CN118311971B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of robot path planning, and more specifically, to a path determination method and a motion control method. Background Art
[0002] With the development of artificial intelligence and micro-robot technology, when patrolling inside oil-immersed power transformers, small patrol robots can be used to locate and analyze transformer internal faults more quickly and intuitively. Due to the huge size of large transformers, the patrol robots have to swim for a long distance and time inside them. Therefore, it is necessary to plan a shorter collision-free path in the transformer, requiring the small patrol robots to swim to the target point while consuming as little energy as possible, so as to efficiently, timely and accurately complete the confirmation and maintenance of faults and defective states inside confined spaces such as transformers.
[0003] In the process of realizing the concept of the present invention, the inventors found that when the related technology uses three-dimensional space path search, the generated path is usually highly random, which leads to problems such as long search time and large algorithm calculation amount. Summary of the invention
[0004] In view of this, the present invention provides a path determination method and a motion control method.
[0005] One aspect of the present invention provides a path determination method, comprising: in response to a path determination request, determining an initial grid of a robot in a grid system and a target grid of a target position included in the path determination request in the grid system; using the range of the enclosed space and the respective positions of the at least one obstacle object as constraints, generating multiple initial paths from the initial grid to the target grid in the enclosed space; performing path optimization based on the multiple initial paths to obtain a target path from the initial grid to the target grid.
[0006] Another aspect of the present invention provides a motion control method, comprising: in response to a motion control request, obtaining a target path from an initial grid where the robot is located in a grid system to a target grid where a target position is located in the above grid system; using a fuzzy algorithm, controlling the above robot to move along the above target path to the above target position; wherein the above target path includes that obtained using any of the above-mentioned path determination methods.
[0007] According to another aspect of the present invention, an electronic device is provided, comprising: one or more processors; and a memory for storing one or more instructions, wherein when the one or more instructions are executed by the one or more processors, the one or more processors implement the method described in the present invention.
[0008] According to another aspect of the present invention, a computer-readable storage medium is provided, on which executable instructions are stored. When the executable instructions are executed by a processor, the processor implements the method according to the present invention.
[0009] According to another aspect of the present invention, a computer program product is provided. The computer program product comprises computer executable instructions. When the computer executable instructions are executed, they are used to implement the method according to the present invention.
[0010] According to an embodiment of the present invention, by simplifying the confined space where the robot is located and the obstacle objects in the confined space into a grid system, and determining the initial grid where the robot is located in the grid system and the target grid where the target position included in the path determination request is located in the grid system, multiple initial paths from the initial grid to the target grid are generated in the confined space based on constraint conditions, thereby simplifying the amount of calculation; at the same time, path optimization is performed on the multiple initial paths to obtain the target path from the initial grid to the target grid, so that the robot can reach the target position in the confined space through the determined optimal target path, thereby improving the accuracy of path planning and reducing the time consumption of path planning. BRIEF DESCRIPTION OF THE DRAWINGS
[0011] The above and other objects, features and advantages of the present invention will become more apparent through the following description of the embodiments of the present invention with reference to the accompanying drawings, in which:
[0012] Figure 1 The application scenarios in which the path determination method and the motion control method according to the embodiments of the present invention can be applied are shown;
[0013] Figure 2 A flow chart of a path determination method according to an embodiment of the present invention is shown;
[0014] Figure 3 A simplified grid diagram of a transformer inside a specific embodiment of the present invention is shown;
[0015] Figure 4 A schematic diagram showing an initial path of a specific embodiment of the present invention;
[0016] Figure 5 A schematic diagram of a gene segment mutation operation according to an embodiment of the present invention is shown;
[0017] Figure 6 A schematic diagram showing a target path according to an embodiment of the present invention is shown;
[0018] Figure 7 A flow chart of a motion control method according to an embodiment of the present invention is shown;
[0019] Figure 8A schematic diagram of a membership function curve of the distance between a robot and an obstacle object according to an embodiment of the present invention is shown;
[0020] Fig. 9 A schematic diagram showing a motion control method according to an embodiment of the present invention is shown;
[0021] Fig.10 A schematic diagram showing a robot detection process according to an embodiment of the present invention is shown;
[0022] Fig.11 A block diagram of a path determination device according to an embodiment of the present invention is shown;
[0023] Fig.12 A block diagram of a motion control device according to an embodiment of the present invention is shown;
[0024] Fig.13 A block diagram of an electronic device suitable for implementing a path determination method and a motion control method according to an embodiment of the present invention is shown. DETAILED DESCRIPTION
[0025] Below, embodiments of the present invention will be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the present invention. In the following detailed description, for ease of explanation, many specific details are set forth to provide a comprehensive understanding of embodiments of the present invention. However, it is apparent that one or more embodiments may also be implemented without these specific details. In addition, in the following description, descriptions of known structures and technologies are omitted to avoid unnecessary confusion of concepts of the present invention.
[0026] The terms used herein are only for describing specific embodiments and are not intended to limit the present invention. The terms "comprise", "include", etc. used herein indicate the existence of the features, steps, operations and / or components, but do not exclude the existence or addition of one or more other features, steps, operations or components.
[0027] All terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art unless otherwise defined.
[0028] When an expression like "at least one of A, B, and C, etc." is used, it should generally be interpreted according to the meaning of the expression that is generally understood by those skilled in the art.
[0029] With the development of artificial intelligence and micro-robot technology, micro-robot fish or bionic robots can be used to more quickly and intuitively locate and analyze internal faults of oil-immersed power transformers during patrol inspections. Since small inspection robots have the advantages of small size and flexible movement, by placing small inspection robots inside the transformer to directly observe the presence and length of carbon marks on the surface of the windings and insulating paperboard, the location of the insulation fault and the degree of insulation degradation can be determined more intuitively.
[0030] Small inspection robots need to patrol in the oil of large transformers. Large transformers are huge in size, and the inspection robots have to swim a long distance and time inside them. The transformer inspection robots are required to swim to the target point while consuming as little energy as possible. Therefore, how to plan a shorter collision-free path in the transformer, so as to efficiently, timely and accurately complete the confirmation and maintenance of faults and defective conditions inside confined spaces such as transformers, is of great significance to the task completion of small inspection robots.
[0031] Spatial path planning is one of the key technologies of intelligent control. The goal is to avoid obstacles and reach the designated location safely according to a certain optimal indicator when the surrounding environment is known or partially known. Compared with intelligent devices moving in two-dimensional space, the complexity and difficulty of path planning in three-dimensional space are greatly increased as the dimension increases. Common algorithms that can be used to solve three-dimensional space trajectory planning problems include A* search method, rapidly expanding random tree, case-based reasoning method, ant colony algorithm, etc., but they all have problems such as large amount of calculation, long planning time, and unsuitability for real-time computing environment. This is mainly manifested in that when searching for paths in three-dimensional space, the generated path has a large randomness, which will result in long search time, slow convergence, and easy deadlock. For path planning in three-dimensional space with complex environment, this shortcoming is more obvious, and it is difficult to find a feasible path that meets the constraints.
[0032] In view of this, the present invention simplifies the confined space where the robot is located and the obstacle objects in the confined space into a grid system, and determines the initial grid where the robot is located in the grid system and the target grid where the target position included in the path determination request is located in the grid system, and generates multiple initial paths from the initial grid to the target grid in the confined space based on constraint conditions, thereby simplifying the amount of calculation; at the same time, path optimization is performed on the multiple initial paths to obtain the target path from the initial grid to the target grid, so that the robot can reach the target position in the confined space through the determined optimal target path, thereby improving the accuracy of path planning and reducing the time consumption of path planning.
[0033] Specifically, an embodiment of the present invention provides a path determination method and a motion control method. The path determination method includes, in response to a path determination request, determining an initial grid in which a robot is located in a grid system and a target grid in which a target position included in the path determination request is located in the grid system, wherein the grid system is determined based on a confined space in which the robot is located, and the confined space includes at least one obstacle object; using the range of the confined space and the respective positions of at least one obstacle object as constraints, generating multiple initial paths from the initial grid to the target grid in the confined space; performing path optimization based on the multiple initial paths to obtain a target path from the initial grid to the target grid.
[0034] It should be noted that the path determination method and motion control method determined in the embodiment of the present invention can be used in the field of robot path planning. The path determination method and motion control method determined in the embodiment of the present invention can also be used in any field other than the field of robot path planning. The application field of the path determination method and motion control method determined in the embodiment of the present invention is not limited.
[0035] In the embodiments of the present invention, the collection, updating, analysis, processing, use, transmission, provision, invention, storage, etc. of the data involved (for example, including but not limited to user personal information) comply with the provisions of relevant laws and regulations, are used for legal purposes, and do not violate public order and good morals.
[0036] In the embodiment of the present invention, before obtaining or collecting the user's personal information, the user's authorization or consent is obtained.
[0037] Figure 1 The following illustrates the application scenarios in which the path determination method and motion control method according to the embodiments of the present invention can be applied. Figure 1 The illustrated examples are merely scenarios to which the embodiments of the present invention may be applied, to help those skilled in the art understand the technical content of the present invention, but do not mean that the embodiments of the present invention may not be applied to other devices, systems, environments or scenarios.
[0038] like Figure 1 As shown, the application scenario according to this embodiment may include a confined space 101 , an obstacle object 102 and a robot 103 .
[0039] The obstacle object 102 and the robot 103 are in the confined space 101 , and the robot 103 needs to find a path from a starting point to a target point in the confined space 101 , wherein the path needs to avoid the obstacle object 102 and the boundary of the confined space 101 .
[0040] It should be understood that Figure 1The number of confined spaces, obstacle objects and robots in the example is only for illustration. Any number of confined spaces, obstacle objects and robots may be provided as required.
[0041] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.
[0042] Figure 2 A flow chart of a path determination method according to an embodiment of the present invention is shown.
[0043] like Figure 2 As shown, the method includes operations S210 to S230.
[0044] In operation S210, in response to a path determination request, an initial grid in which the robot is located in a grid system and a target grid in which a target position included in the path determination request is located in the grid system are determined.
[0045] According to an embodiment of the present invention, the grid system is determined based on the confined space where the robot is located. Specifically, a preset coordinate system can be established in the confined space where the robot is located, wherein the preset coordinate system can be a spatial Cartesian coordinate system established with one end point of the confined space as the origin and three mutually perpendicular sides of the confined space as coordinate axes.
[0046] According to an embodiment of the present invention, after a coordinate system is established in a closed space, three coordinate axes are evenly divided into equal lengths, and the closed space is grid-modeled to obtain a grid system composed of a plurality of grids.
[0047] According to an embodiment of the present invention, the path determination request may be used to determine the path of the robot from the initial grid to the target grid in the confined space, wherein the initial grid of the robot may be determined by the positioning signal of the robot itself.
[0048] According to an embodiment of the present invention, the target position may be a position in a confined space that the robot needs to reach, and a target grid may be determined in a preset coordinate system according to the target position.
[0049] According to an embodiment of the present invention, the enclosed space includes at least one obstacle object. For example, one or more windings are arranged in an enclosed metal shell of a transformer, and the one or more windings can serve as obstacle objects.
[0050] In operation S220, a plurality of initial paths from the initial grid to the target grid are generated in the enclosed space with the range of the enclosed space and the respective positions of at least one obstacle object as constraints.
[0051] According to an embodiment of the present invention, after determining the initial grid and target grid of the robot in a confined space, the grids corresponding to the interior of the confined space and the positions of one or more obstacle objects are used as constraints, and multiple initial paths are generated in the confined space, wherein the initial paths can be used to characterize the path from the initial grid corresponding to the robot's position to the target grid corresponding to the target position.
[0052] For example, after simplifying the transformer into a grid system and determining the initial grid and target grid of the robot in the transformer, within the internal space of the metal shell of the transformer where the robot can move, the grids corresponding to the positions of one or more obstacle objects such as windings are used as constraints to generate multiple initial paths from the initial grid corresponding to the robot's position to the target grid corresponding to the target position, so that the robot can reach the target grid while avoiding obstacle objects in the transformer.
[0053] In operation S230 , path optimization is performed based on the multiple initial paths to obtain a target path from the initial grid to the target grid.
[0054] According to an embodiment of the present invention, after a plurality of initial paths are determined, path optimization may be performed on the plurality of initial paths to obtain an optimal path from the initial grid to the target grid as the target path.
[0055] Figure 3 A simplified internal grid schematic diagram of a transformer according to a specific embodiment of the present invention is shown.
[0056] like Figure 3 As shown, in a closed metal shell of a transformer, there are one or more windings, and a spatial Cartesian coordinate system O-XYZ is established in the transformer, wherein the coordinate origin is set at the lower left corner of the transformer, and Xmax, Ymax and Zmax are the maximum values on the X, Y and Z axes respectively. The three coordinate axes are evenly divided into equal lengths, and the transformer is modeled by a 60×30×10 grid. When the components in the transformer do not occupy more than one grid, they can be calculated as occupying one grid, thereby obtaining a transformer after grid simplification. The simplified transformer can be used as a grid system, and one or more windings inside can be used as obstacle objects, thereby realizing the simplification of the transformer.
[0057] An embodiment of the present invention simplifies the confined space where the robot is located and the obstacle objects in the confined space into a grid system, determines the initial grid where the robot is located in the grid system and the target grid where the target position included in the path determination request is located in the grid system, generates multiple initial paths from the initial grid to the target grid in the confined space based on constraint conditions, and optimizes the multiple initial paths to obtain a target path from the initial grid to the target grid, so that the robot can reach the target position in the confined space through the determined optimal target path, thereby improving the accuracy of path planning and reducing energy consumption.
[0058] According to an embodiment of the present invention, a plurality of initial paths from an initial grid to a target grid are generated in the confined space with the range of the confined space and the respective positions of at least one obstacle object as constraints, including: taking the target grid as a gravitational source, taking the grid where the inner wall of the confined space is located and the grid where at least one obstacle object is located as repulsive sources, to obtain a potential field in the confined space; in the process of generating each initial path, using an artificial potential field method, to determine a plurality of intermediate grids from a grid system based on the potential field; and obtaining an initial path based on the initial grid, the target grid and the plurality of intermediate grids.
[0059] According to an embodiment of the present invention, when the range of the confined space and the respective positions of one or more obstacle objects are used as constraints, the motion space inside the transformer where the robot is located can be first defined as an abstract potential field, which can be composed of a gravitational field and a repulsive field. The potential field can be defined as the superposition of the gravitational field of the target grid and the repulsive field of the grid where the inner wall of the confined space is located and the grid where at least one obstacle object is located.
[0060] According to an embodiment of the present invention, the gravitational field is a potential field generated on the robot by taking the target grid as a gravitational source. The direction of the gravitational field can be defined as pointing from the grid where the robot is located to the target grid, and the gravitational field increases monotonically as the distance between the robot and the target grid increases. For example, the farther the distance between the robot and the target grid is, the greater the gravitational force it receives.
[0061] According to an embodiment of the present invention, the repulsive field is a potential field generated on the robot by using the grid where the inner wall of the confined space is located and the grid where at least one obstacle object is located as the repulsive source. The direction of the repulsive field can be defined as pointing to the grid where the inner wall of the confined space is located and the grid where at least one obstacle object is located, and the repulsive field has a maximum value when the obstacle object is located in the grid, and decreases monotonically with the increase of the distance between the robot and the obstacle object or the distance from the inner wall of the confined space. For example, the farther the distance between the robot and the obstacle object or the inner wall of the confined space is, the smaller the repulsive force is.
[0062] According to an embodiment of the present invention, the artificial potential field method controls the movement of the robot through the combined force of gravity and repulsion, so that the robot effectively avoids obstacle objects and boundaries of confined spaces and moves to a target grid position.
[0063] According to an embodiment of the present invention, in the process of generating each initial path, after determining the initial grid and the target grid, multiple intermediate grids can be determined from the grid system based on the artificial potential field method. The intermediate grids can be used to represent the grids that the robot passes through from the initial grid to the target grid.
[0064] According to an embodiment of the present invention, after an initial grid, a target grid and a plurality of intermediate grids in a confined space are determined, a plurality of initial paths from the initial grid through the intermediate grids to the target grid may be generated in the confined space.
[0065] According to an embodiment of the present invention, an artificial potential field method is used to determine multiple intermediate grids from a grid system based on the potential field, including: for the ith intermediate grid, using the artificial potential field method to determine the potential field forces of the ith intermediate grid in the potential field; based on the potential field forces and the movement directions of the ith intermediate grid to each of the multiple adjacent grids, determining the transfer probability from the ith intermediate grid to each of the multiple adjacent grids; based on the transfer probability from the ith intermediate grid to each of the multiple adjacent grids, selecting the i+1th intermediate grid from the multiple adjacent grids.
[0066] According to an embodiment of the present invention, the potential field stress of the i-th intermediate grid in the potential field is calculated as follows.
[0067] (1);
[0068] (2);
[0069] (3);
[0070] In the formula, F to (P f ) is the potential field force, F at (P f ) represents the gravitational force, F rej (P f ) is expressed as repulsive force, K at Expressed as the gravitational coefficient, K re Expressed as the repulsion coefficient, P f Represents the grid position of the robot, P oj Represents the grid position of obstacle object j, P f— P oj It is represented as the distance between the robot and the obstacle object j, It is represented as the maximum impact distance of the obstacle object, x, y, and z are the grid coordinates of the robot. Represented as the potential field vector on the x-axis coordinate of the robot, Represented as the potential field vector on the y-axis coordinate of the robot, Represented as the potential field vector on the y-axis coordinate of the robot, x oj ,y oj 、z oj The obstacle object j is represented by its grid coordinates, x g ,y g 、z g Respectively represented as target grid coordinates.
[0071] According to an embodiment of the present invention, after determining the potential field forces of the i-th intermediate grid in the potential field, based on the potential field forces and the movement directions of the i-th intermediate grid to each of the multiple adjacent grids, the transfer probability from the i-th intermediate grid to each of the multiple adjacent grids is determined by calculating the angular difference between the movement direction of the grid occupied by the robot to the adjacent grid and the potential field forces, wherein the multiple adjacent grids are all adjacent to the i-th intermediate grid, and the transfer probability of the adjacent grids is calculated as follows.
[0072] (4);
[0073] In the formula, p ij It is expressed as the transition probability from the i-th intermediate grid to the j-th intermediate grid, α ij It is represented as the angle between the resultant force and the direction of the jth intermediate grid, V is represented as the set of adjacent grids around the robot, k is represented as the kth adjacent grid around the robot, α ik It is expressed as the angle between the resultant force and the direction of the kth adjacent grid.
[0074] According to an embodiment of the present invention, after calculating the transition probability p of each neighborhood grid being selected as the next moving grid, ij , the next grid can be selected from multiple adjacent grids and used as the intermediate grid.
[0075] For example, the selected grid can be marked as an obstacle grid, and the selected grid can be used as the current grid. Determine whether the current grid has reached the target point. If not, recalculate the angle difference between the movement direction of the grid occupied by the robot to the adjacent grid and the potential field force, and then recalculate the transition probability p of the adjacent grid. ij ; If the target point is reached, the current path is saved and used as the initial path.
[0076] Reference below Figure 4 , the method for determining the initial path is further explained in conjunction with a specific embodiment.
[0077] Figure 4 A schematic diagram of an initial path according to a specific embodiment of the present invention is shown.
[0078] like Figure 4 As shown, in the scenario where the robot detects the inside of the transformer, the schematic diagram includes a transformer housing 401, a winding 402, a starting point 403, a target point 404 and an initial path 405. The transformer is gridded to obtain a transformer grid system, and the grids where the transformer housing 401, the winding 402, the starting point 403 and the target point 404 are located are determined. The grid where the starting point 403 is located is used as the initial grid, and the grid where the target point 404 is located is used as the target grid. Based on the internal potential field of the transformer, multiple initial paths 405 are determined, so that the robot reaches the target point 404 from the starting point 403 while avoiding the winding 402 and the inner wall of the transformer housing 401, wherein the grids where the multiple initial paths 405 are located are used as intermediate grids.
[0079] The embodiments of the present invention establish gravitational fields and repulsive fields in a grid system, and calculate the angular difference between the movement direction of the robot from the grid occupied by the robot to the adjacent grid and the potential field force through the artificial potential field method, and determine multiple intermediate grids based on the adjacent grid movement probability formula, thereby realizing the rapid creation of multiple initial paths and improving the efficiency and accuracy of path planning.
[0080] According to an embodiment of the present invention, path optimization is performed based on multiple initial paths to obtain a target path from an initial grid to a target grid, including: using the multiple initial paths as multiple initial individuals, performing multiple rounds of crossover and mutation operations on the multiple initial individuals using a genetic algorithm to obtain multiple target individuals; and obtaining a target path based on the multiple target individuals.
[0081] According to an embodiment of the present invention, when the number of initial paths reaches a set number, the multiple initial paths obtained can be used as initial individuals, and multiple rounds of selection, crossover, and mutation operations can be performed on the multiple initial individuals using a genetic algorithm to obtain multiple target individuals processed by the genetic algorithm. Based on the multiple target individuals, the optimal target individual is selected as the target path.
[0082] According to an embodiment of the present invention, multiple initial paths are used as multiple initial individuals, and multiple rounds of crossover and mutation operations are performed on the multiple initial individuals using a genetic algorithm to obtain multiple target individuals, including: in the j-th round of crossover and mutation operation process, multiple first current individuals are obtained based on the population corresponding to the j-th round of crossover and mutation operation process; the fitness of each of the multiple first current individuals is determined based on the path information of the paths corresponding to each of the multiple first current individuals; based on the fitness of each of the multiple first current individuals, a crossover operation is performed on the multiple first current individuals to obtain multiple second current individuals; a mutation operation is performed on the multiple second current individuals to obtain multiple third current individuals; and a preset number of first current individuals in the multiple first current individuals are used to replace a preset number of third current individuals in the multiple third current individuals to obtain a population corresponding to the j+1-th round of crossover and mutation operation process.
[0083] According to an embodiment of the present invention, j is a positive integer, N represents a crossover mutation operation round number threshold, and when j=1, multiple initial individuals can be used as multiple first current individuals.
[0084] According to an embodiment of the present invention, in a genetic algorithm, a fitness function can be used to evaluate multiple initial individuals, i.e., first current individuals, so that the fitness of each of the multiple first current individuals can be determined. The fitness function can be determined by considering factors such as path length, a penalty factor for the distance between the path and the obstacle object, the maximum turning angle of the path, and the total turning angle of the path. The fitness function is constructed as follows.
[0085] (5);
[0086] In the formula, f(p) represents the fitness function corresponding to the first current individual, d i It is represented as the length of the i-th segment of the path, A i Expressed as the distance penalty factor between the path and the obstacle object, θ i Represented as the i-th turning angle of the path, θ max It is represented as the maximum turning angle in the path, n is represented as the number of road sections, and w1, w2, w3, and w4 are represented as weight coefficients.
[0087] According to an embodiment of the present invention, the penalty factor of the distance between a path and an obstacle object is calculated as follows.
[0088] (6);
[0089] In the formula, γ1 represents the collision distance between the robot and the nearest obstacle object, γ2 represents the ideal distance between the robot and the nearest obstacle object, and e i Indicates the actual distance between the robot and the nearest obstacle object.
[0090] According to an embodiment of the present invention, based on the calculated fitness of each of the multiple first current individuals, a crossover operation can be performed on the multiple first current individuals, and the individual after the crossover operation can be used as the second current individual; a mutation operation can be performed on the multiple second current individuals, and the individual after the compilation operation can be used as the third current individual.
[0091] A preset number of first current individuals among the multiple first current individuals are used to replace a preset number of third current individuals among the multiple third current individuals to obtain a population corresponding to the next round of crossover and mutation operation process, wherein the population corresponding to the next round of crossover and mutation operation process includes multiple target individuals.
[0092] According to an embodiment of the present invention, when the third current individual is determined, a preset number of first current individuals can also be used to replace a plurality of preset number of third current individuals based on a random selection strategy. The random selection strategy may include a roulette method, a binary tournament method, an exponential sorting method, etc. For example, in an embodiment of the present invention, a nonlinear ranking selection strategy based on a roulette method and an optimal retention strategy can preferably be adopted to select the top 10% of the fitness rankings in the first current individuals to replace the third current individuals with lower fitness rankings in the bottom 10% generated by operations such as crossover and mutation in the current generation population, so as to prevent them from being destroyed in subsequent crossover and mutation operations.
[0093] The embodiment of the present invention uses a genetic algorithm to perform computational processing on the first current individual, and based on the fitness function of the selected path, selects a shorter path, a maximum turning angle that satisfies the robot's turning radius, a smaller path turning angle, and a collision-free optimal path, thereby meeting the robot's path planning requirements in the complex environment inside a confined space.
[0094] According to an embodiment of the present invention, based on the respective fitness of the plurality of first current individuals, a crossover operation is performed on the plurality of first current individuals to obtain a plurality of second current individuals, including: determining a plurality of individual groups based on the plurality of first current individuals; determining a selection probability of the plurality of individual groups based on the respective fitness of the plurality of first current individuals; determining a plurality of target individual groups from the plurality of individual groups based on the respective selection probabilities of the plurality of individual groups; and performing a crossover operation on two first current individuals included in each of the plurality of target individual groups to obtain a plurality of second current individuals.
[0095] According to an embodiment of the present invention, after a plurality of target first current individuals are determined, a plurality of individual groups are determined from the plurality of target first current individuals based on a crossover algorithm, wherein each individual group includes two parent individuals for crossover.
[0096] According to an embodiment of the present invention, in order to enable the crossover operation between two excellent parents to produce possibly better offspring, the probability of an individual group being selected from a population in an embodiment of the present invention is affected by its respective adaptability. The higher the fitness ranking of the first current individual, the greater the probability of being selected as a parent. Therefore, the selection probability of each of the multiple individual groups can be determined based on the respective fitness of the multiple first current individuals. The selection probability of the individual group is calculated as follows.
[0097] (7);
[0098] Where P c Expressed as the selection probability of an individual group, f max (p) represents the maximum fitness value of the first current individual, f1(p1) is the fitness value of the first current individual p1, and f2(p2) is the fitness value of the first current individual p2.
[0099] According to an embodiment of the present invention, when determining the selection probability of each of multiple individual groups, multiple target individual groups are determined from the multiple individual groups based on the calculated selection probabilities of each of the multiple individual groups and preset rules. For example, multiple individual groups with larger selection probabilities can be used as target individual groups.
[0100] According to an embodiment of the present invention, a crossover operation is performed on two first current individuals respectively included in a plurality of target individual groups to obtain a plurality of second current individuals, including: for each target individual group, when there is an identical grid in two paths respectively corresponding to the two first current individuals included in the individual group, a crossover operation is performed on the paths respectively corresponding to the two first current individuals included in the individual group based on the identical grid to obtain two second current individuals.
[0101] According to an embodiment of the present invention, the above-mentioned crossover operation can adopt the overlapping point crossover method, and select points with the same grid number for each target individual group for crossover operation. When the same grid exists in two paths corresponding to the two first current individuals included in the individual group, one of them is randomly selected for crossover to obtain the second current individual.
[0102] According to an embodiment of the present invention, when there is no same grid in two paths respectively corresponding to two first current individuals included in the individual group, the individual group is reselected to perform the crossover operation.
[0103] An embodiment of the present invention determines multiple target individual groups from multiple individual groups by determining the selection probability of each of the multiple individual groups. Since the probability of an individual group being selected from a population is affected by its respective adaptability, the individual group with a higher fitness is used as the target individual group for crossover operation, which can improve the accuracy of the second current individual and thus increase the probability of obtaining the optimal path.
[0104] According to an embodiment of the present invention, a mutation operation is performed on multiple second current individuals to obtain multiple third current individuals, including: for each second current individual, based on the path corresponding to each second current individual, determining multiple path segments included in each second current individual; selecting a first target path segment from the multiple path segments included in each second current individual; generating a second target path segment based on two path segment nodes corresponding to the first target path segment; and when the second target path segment does not intersect with the grids where at least one obstacle object is located, using the second target path segment to replace the first target path segment included in each second current individual to obtain the third current individual.
[0105] According to an embodiment of the present invention, when multiple second current individuals are determined, the paths corresponding to the multiple second current individuals are evenly divided into multiple gene segments. From the multiple path segments included in each second current individual, a gene segment is randomly selected as the first target path segment.
[0106] According to an embodiment of the present invention, based on two path segment nodes corresponding to the first target path segment, a gene mutation operation is performed on the path segment nodes, so that each node in the first target path segment is on the same straight line and the length of the gene segment is shortened, and the second target path segment is generated after mutation.
[0107] According to an embodiment of the present invention, when the second target path segment does not intersect with the grids in which at least one obstacle object is located, it means that the gene mutation process can be used as an effective gene mutation, and the second target path segment can be used to replace the first target path segment included in each second current individual to obtain a third current individual after replacement.
[0108] According to an embodiment of the present invention, when the second target path segment intersects with the grid where at least one obstacle object is located, the current gene mutation is abandoned and other gene segments are reselected for a new round of gene mutation.
[0109] Reference below Figure 5 , the method of gene fragment variation operation is further explained in conjunction with a specific embodiment.
[0110] Figure 5 A schematic diagram of a gene segment mutation operation according to an embodiment of the present invention is shown.
[0111] like Figure 5As shown, the original gene segment and the second target path segment obtained after gene mutation have the same gene segment starting point and gene segment end point. The original gene segment is curved, and the second target path segment obtained after gene mutation is straight, and each node is on the same straight line, and the length of the second target path segment is shortened, thereby obtaining a target path with a shorter path.
[0112] Reference below Figure 6 , the target path obtained after the genetic algorithm iteration is further explained in conjunction with a specific embodiment.
[0113] Figure 6 A schematic diagram of a target path according to an embodiment of the present invention is shown.
[0114] like Figure 6 As shown in FIG. 4 , in a scenario where the robot detects the inside of a transformer, the schematic diagram includes a transformer housing 401, a winding 402, a starting point 403, a target point 404, and a target path 4051. After obtaining multiple initial paths by the artificial potential field method, the multiple initial paths are used as multiple initial individuals, and the genetic algorithm is used to perform multiple rounds of selection, crossover, and mutation operations on the multiple initial individuals to obtain multiple target individuals. For example, the multiple initial individuals are divided into groups of two to obtain multiple individual groups, and the selection probability P of the individual group being selected as the crossover is calculated. c Set it to 0.8 and change the mutation probability P m Set to 0.1, and after iterating the selected individual group 100 times, a shortest smooth collision-free path is obtained, that is, Figure 6 The target path 4051 in is used as the global planning path.
[0115] The embodiment of the present invention performs selection, crossover and mutation operations on the initial path based on a genetic algorithm, divides the path corresponding to the second current individual into multiple path segments on average, randomly selects a path segment and performs a genetic mutation operation on it, so that each node in the path segment is on the same straight line and the length of the path segment is shortened, thereby obtaining a target path with a shorter path by performing crossover and mutation on the initial path, thereby completing the optimization of the target path.
[0116] Figure 7 A flow chart of a motion control method according to an embodiment of the present invention is shown.
[0117] like Figure 7 As shown, the motion control method includes S710 and S720.
[0118] In operation S710, in response to a motion control request, a target path from an initial grid where the robot is located in a grid system to a target grid where a target position is located in the grid system is acquired.
[0119] According to an embodiment of the present invention, the motion control request may be used to perform local path planning for a target path in a confined space, so that the robot can effectively avoid obstacles when moving.
[0120] According to an embodiment of the present invention, in response to receiving a motion control request, a target path is obtained from the initial grid where the robot is located in the grid system to the target grid where the target position is located in the grid system. The target path can be obtained based on the above-mentioned path determination method S210~S230.
[0121] In operation S720, the robot is controlled to move to a target position along a target path using a fuzzy algorithm.
[0122] For example, when the robot's target path is obtained, a fuzzy algorithm can be used to control the robot to move along the target path to the target position. Fuzzy control can be implemented by a fuzzy controller, which can include four parts: fuzzification, fuzzy rules, fuzzy reasoning, and anti-fuzzy interface. The input of the fuzzy controller can include the distance and angle between the robot and the obstacle object, and the output can include the angle of rotation of the robot when it moves.
[0123] The embodiment of the present invention receives a motion control request, and based on the acquired target path and the distance and angle between the robot and the obstacle object, continuously changes the rotation angle of the robot based on a fuzzy control algorithm, thereby effectively avoiding the obstacle and realizing local path planning of the robot.
[0124] According to an embodiment of the present invention, the target path includes multiple path segments formed based on multiple path nodes; wherein, the use of a fuzzy algorithm to control the robot to move along the target path to the target position includes: when controlling the robot to move to the kth path node, obtaining obstacle information of the kth path segment through the robot; when the obstacle information of the kth path segment indicates that there is an obstacle in the kth path segment, using a fuzzy algorithm to perform local path planning based on the kth path node and the k+1th path node to obtain the target path; controlling the robot to move along the target path to the k+1th path node.
[0125] According to an embodiment of the present invention, the target path includes multiple path segments formed based on multiple path nodes. When the robot is controlled to move to the kth path node, the robot obtains obstacle information of the kth path segment, wherein the k path segments represent the path between the kth path node and the k+1th path node in the target path, and the obstacle information of the kth path segment may include obstacle object information within the robot's ranging range, such as the distance and angle between the robot and the obstacle object, which can be obtained by sensors such as a ranging module and a visual camera installed on the top of the robot.
[0126] According to an embodiment of the present invention, if the obstacle information of the kth path segment indicates that there is an obstacle in the path segment, a fuzzy algorithm can be used to perform local path planning based on the kth path node and the k+1th path node to obtain the target path.
[0127] According to an embodiment of the present invention, firstly, a fuzzy language description is performed on the input variables and the output variables, wherein the input variables may include the distance d between the robot and the obstacle object, the angle Φ between the robot and the obstacle object, and the output variables may include the adjustment angle α of the robot.
[0128] For example, the range of the distance measured by the distance sensor at the end of the robot can be uniformly quantized to [0, 8], the domain of the distance d between the robot and the obstacle object is [0, 2, 4, 6, 8], and the corresponding fuzzy set is d={VC, C, M, F, VF}={"negative large", "negative small", "zero", "positive small", "positive large"}. At the same time, the range of the angle Φ between the input robot and the obstacle object is non-uniformly quantized to the interval [-4, 4]. For example, it can be uniformly quantized in the interval [-65°, 65°]. When Φ>65°, it is quantized to 4, and when Φ<-65°, it is quantized to -4. The domain of Φ is [-4, -2, 0, 2, 4], and the corresponding fuzzy set is Φ={RB, RS, Z, LS, LB}={"right large", "right small", "zero", "left small", "left large"}. The range of the output robot's rotation angle α [-65°, 65°] is uniformly quantized to the interval [-4, 4]. The domain is the same as Φ, and the corresponding fuzzy set of the output variable α is V={ORB, ORS, OZ, OLS, OLB}=={"right large", "right small", "zero", "left small", "left large"}.
[0129] According to the embodiment of the present invention, a knowledge base is established. The knowledge base mainly consists of a database and a rule base. The database may include the membership function of each linguistic variable, the scale transformation factor, and the number of fuzzy space classifications, etc. The rule base may include a series of control rules represented by fuzzy linguistic variables, which is composed of the sum of several fuzzy "if...then". The membership function μ(d) of the fuzzy set can be determined by a normal distribution function, and the calculation method of the membership function is as follows.
[0130] (8);
[0131] Where μ(d) represents the membership function of the distance between the robot and the obstacle object, c represents the fuzzy set parameter, σ represents the standard deviation of the normal distribution, which can be 1, and d represents the distance between the robot and the obstacle object.
[0132] According to an embodiment of the present invention, the fuzzy set parameter c of the distance d between the robot and the obstacle object can be 0, 2, 4, 6, 8 respectively, the fuzzy set parameter c of the angle Φ between the robot and the obstacle object can be -4, -2, 0, 2, 4 respectively, and the fuzzy set parameter c of the adjustment angle α of the output end robot can be -4, -2, 0, 2, 4 respectively.
[0133] Figure 8 A schematic diagram of a membership function curve of the distance between a robot and an obstacle object according to an embodiment of the present invention is shown.
[0134] like Figure 8 As shown, the membership function is a Gaussian membership function, in which "negative large" can be expressed as a large deviation in the negative direction, which can be expressed as VC, "negative small" can be expressed as a small deviation in the negative direction, which can be expressed as C, "zero" can be expressed as a deviation close to zero, which can be expressed as M, "positive small" can be expressed as a small deviation in the positive direction, which can be expressed as F, and "positive large" can be expressed as a large deviation in the positive direction, which can be expressed as VF.
[0135] According to an embodiment of the present invention, the fuzzy controller has two input variables, d and Φ, and an output variable, α, wherein the control rule of the fuzzy controller can be expressed by a fuzzy statement "If d and Φ then α". Specifically, the control rules of the system are summarized as shown in Table 1 below.
[0136] Table 1
[0137]
[0138] According to an embodiment of the present invention, fuzzy reasoning can adopt linguistic fuzzy reasoning (Mamdani), which has the ability to simulate human reasoning based on fuzzy concepts and is based on the implication relationship and reasoning rules in fuzzy logic. According to the input of the fuzzy system and the control rules in the rule base, after logical operations such as fuzzy relationship synthesis and fuzzy reasoning synthesis, a conclusion is derived from some fuzzy premise conditions, thereby obtaining the output of the fuzzy system.
[0139] According to an embodiment of the present invention, after obtaining the fuzzy reasoning result through fuzzy reasoning, it also includes defuzzifying the fuzzy reasoning result, that is, converting the fuzzy reasoning result into an accurate value. Among them, the defuzzification method may include the area centroid method, the area equal division method, the average maximum membership method, the maximum membership method, the maximum membership method, etc. The embodiment of the present invention can use the centroid method to defuzzify the fuzzy reasoning result, and the calculation method of the defuzzification method is as follows.
[0140] (9);
[0141] Where α0 is the clarity value obtained after defuzzification by the centroid method. It is expressed as the algebraic integral of the membership of all elements of the output fuzzy subset on the continuous domain, α l It indicates the adjustment angle of the output corresponding to the maximum value of the membership function corresponding to each fuzzy subset at the output end.
[0142] Reference below Fig. 9 and Fig.10 , the robot motion control method is further explained in conjunction with a specific embodiment.
[0143] Fig. 9 A schematic diagram of a motion control method according to an embodiment of the present invention is shown.
[0144] like Fig. 9 As shown, when the fuzzy algorithm is used to control the robot to move along the target path to the target position, the control method includes S901~S907.
[0145] In response to receiving a motion control request and obtaining the robot's target path, operation S901 is performed to input the distance between the robot and the obstacle object into the fuzzy controller, and operation S902 is performed to input the angle between the robot and the obstacle object into the fuzzy controller, and then operation S903 is performed to determine the database and rule base in the knowledge base.
[0146] When the fuzzy controller receives the input of the distance and angle between the robot and the obstacle object, the knowledge base is called and operation S904 is performed to fuzzify the input based on the control rules of the rule base in the knowledge base and the content of the database to obtain a fuzzy set. Based on the fuzzy set, operation S905 is performed to perform fuzzy reasoning on the fuzzy set, and then operation S906 is performed to defuzzify the fuzzy reasoning result, and the adjustment angle of the robot is output through operation S907, wherein operations S903 to S906 can be performed by the fuzzy controller.
[0147] According to an embodiment of the present invention, based on the values of the input variables d and Φ, after the robot rotates according to the output of the fuzzy control controller, i.e., the robot's adjustment angle α, it does not immediately move forward by a fixed step length. Since the robot does not know the obstacle information in the new forward direction after the rotation, if it moves forward immediately, it may hit the obstacle. Therefore, after the robot rotates, it also needs to detect whether the distance of the obstacle in front is greater than the fixed step length. If the distance of the obstacle in front is greater than the fixed step length, the robot is controlled to move forward. If the distance of the obstacle in front is not greater than the fixed step length, the robot cannot move forward. Therefore, it is necessary to continue to detect by continuously deflecting the angle in a certain direction (such as to the right) until a safe forward direction is found. At this time, the calculation method of the angle that the robot finally needs to deflect is as follows.
[0148] (10);
[0149] Where τ represents the final deflection angle of the robot, σ i It is expressed as the continuous deflection angle. Among them, when the robot continues to turn left, σ i is positive, when turning right i is negative.
[0150] Fig.10 A schematic diagram of a robot detection process according to an embodiment of the present invention is shown.
[0151] like Fig.10 As shown in the figure, after the robot adjusts the angle to α, if it detects that the distance of the obstacle in front is not greater than the fixed step length, the robot needs to continue detecting at a continuous deflection angle σ until it avoids the obstacle. At this time, the final deflection angle of the robot is τ.
[0152] The embodiments of the present invention address the problem of collision with smaller components or other unconsidered obstacles inside a confined space by adopting a fuzzy algorithm. The robot detects the distance and angle of the obstacle through sensors such as a ranging module at the end of the robot and a visual camera, and continuously changes the swimming direction of the robot based on the fuzzy control algorithm, thereby effectively avoiding the obstacle.
[0153] It should be noted that, unless it is explicitly stated that there is a sequence of execution between different operations shown in the flowchart in the embodiments of the present invention, or different operations have a sequence of execution in technical implementation, otherwise, the execution order of multiple operations may not be particular, and multiple operations may also be executed simultaneously.
[0154] Fig.11 A block diagram of a path determination device according to an embodiment of the present invention is shown.
[0155] like Fig.11 As shown, the path determination device 1100 includes a first determination module 1110 , a path generation module 1120 , and a second determination module 1130 .
[0156] The first determination module 1110 is used to determine, in response to a path determination request, an initial grid where the robot is located in the grid system and a target grid where a target position included in the path determination request is located in the grid system.
[0157] The path generation module 1120 is used to generate multiple initial paths from the initial grid to the target grid in the enclosed space by taking the range of the enclosed space and the position of at least one obstacle object as constraints.
[0158] The second determination module 1130 is used to perform path optimization based on multiple initial paths to obtain a target path from the initial grid to the target grid.
[0159] Fig.12 A block diagram of a motion control device according to an embodiment of the present invention is shown.
[0160] like Fig.12 As shown, the motion control device 1200 includes an acquisition module 1210 and a control module 1220 .
[0161] The acquisition module 1210 is used to acquire a target path from an initial grid where the robot is located in the grid system to a target grid where a target position is located in the grid system in response to a motion control request.
[0162] The control module 1220 is used to control the robot to move to the target position along the target path using a fuzzy algorithm.
[0163] It should be noted that the path determination device and motion control device parts in the embodiments of the present invention correspond to the path determination method and motion control method parts in the embodiments of the present invention. The description of the path determination device and motion control device parts specifically refers to the path determination method and motion control method parts, which will not be repeated here.
[0164] Fig.13A block diagram of an electronic device suitable for implementing a path determination method and a motion control method according to an embodiment of the present invention is shown. Fig.13 The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.
[0165] like Fig.13 As shown, the computer electronic device according to an embodiment of the present invention includes a processor 1301, which can perform various appropriate actions and processes according to the program stored in the read-only memory ROM 1302 or the program loaded from the storage part 1308 to the random access memory RAM 1303. The processor 1301 may include, for example, a general-purpose microprocessor (such as a CPU), an instruction set processor and / or a related chipset and / or a special-purpose microprocessor (for example, an application-specific integrated circuit (ASIC)), etc. The processor 1301 may also include an on-board memory for caching purposes. The processor 1301 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of the present invention.
[0166] In RAM 1303, various programs and data required for the operation of the electronic device are stored. The processor 1301, ROM 1302, and RAM 1303 are connected to each other via a bus 1304. The processor 1301 performs various operations of the method flow according to the embodiment of the present invention by executing the programs in ROM 1302 and / or RAM 1303. It should be noted that the program may also be stored in one or more memories other than ROM 1302 and RAM 1303. The processor 1301 may also perform various operations of the method flow according to the embodiment of the present invention by executing the programs stored in the one or more memories.
[0167] According to an embodiment of the present invention, the electronic device may further include an input / output (I / O) interface 1305, which is also connected to the bus 1304. The electronic device may further include one or more of the following components connected to the input / output (I / O) interface 1305: an input portion 1306 including a keyboard, a mouse, etc.; an output portion 1307 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage portion 1308 including a hard disk, etc.; and a communication portion 1309 including a network interface card such as a LAN card, a modem, etc. The communication portion 1309 performs communication processing via a network such as the Internet. A drive 1310 is also connected to the input / output (I / O) interface 1305 as needed. A removable medium 1311, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 1310 as needed, so that a computer program read therefrom is installed into the storage portion 1308 as needed.
[0168] The present invention further provides a computer-readable storage medium. The computer-readable storage medium carries one or more programs, and when the one or more programs are executed, the method according to the embodiment of the present invention is implemented.
[0169] An embodiment of the present invention also includes a computer program product, which includes a computer program, which contains program code for executing the method provided by the embodiment of the present invention. When the computer program product runs on an electronic device, the program code is used to enable the electronic device to implement the path determination method and motion control method provided by the embodiment of the present invention.
[0170] It will be appreciated by those skilled in the art that the features described in the various embodiments and / or claims of the present invention may be combined and / or coupled in various ways, even if such combinations and / or couplings are not explicitly described in the present invention. In particular, the features described in the various embodiments and / or claims of the present invention may be combined and / or coupled in various ways without departing from the spirit and teachings of the present invention. All of these combinations and / or couplings fall within the scope of the present invention.
[0171] The embodiments of the present invention are described above. However, these embodiments are only for the purpose of illustration, and are not intended to limit the scope of the present invention. Although each embodiment is described above, it does not mean that the measures in each embodiment cannot be used in combination. The scope of the present invention is defined by the attached claims and their equivalents. Without departing from the scope of the present invention, those skilled in the art may make various substitutions and modifications, which should all fall within the scope of the present invention.
Claims
1. A path determination method, It is characterized in that include: In response to a path determination request, determining an initial grid of the robot in a grid system and a target grid of a target position included in the path determination request in the grid system, wherein the grid system is determined based on a confined space where the robot is located, and the confined space includes at least one obstacle object; Using the range of the enclosed space and the respective positions of the at least one obstacle object as constraints, generating a plurality of initial paths from the initial grid to the target grid in the enclosed space; Performing path optimization based on the multiple initial paths to obtain a target path from the initial grid to the target grid includes: Using the multiple initial paths as multiple initial individuals, performing multiple rounds of crossover and mutation operations on the multiple initial individuals using a genetic algorithm to obtain multiple target individuals; Based on the multiple target individuals, obtaining the target path; The multiple initial paths are used as multiple initial individuals, and multiple rounds of crossover and mutation operations are performed on the multiple initial individuals using a genetic algorithm to obtain multiple target individuals, including: In the j-th round of crossover and mutation operation process, a plurality of first current individuals are obtained based on the population corresponding to the j-th round of crossover and mutation operation process, wherein j is a positive integer, j≤N, N represents a crossover and mutation operation round number threshold, and when j=1, the plurality of first current individuals are the plurality of initial individuals; Determining the fitness of each of the plurality of first current individuals based on path information of paths corresponding to each of the plurality of first current individuals; Based on the fitness of each of the multiple first current individuals, performing a crossover operation on the multiple first current individuals to obtain multiple second current individuals; Performing mutation operation on the plurality of second current individuals to obtain a plurality of third current individuals; Replacing a preset number of third current individuals among the plurality of third current individuals with a preset number of first current individuals among the plurality of first current individuals, to obtain a population corresponding to the j+1th round of crossover and mutation operation process, wherein, in the case of j=N, the population corresponding to the j+1th round of crossover and mutation operation process includes the plurality of target individuals; The step of performing mutation operation on the plurality of second current individuals to obtain a plurality of third current individuals includes: For each second current individual, based on the path corresponding to each second current individual, determine a plurality of path segments included in each second current individual; Selecting a first target path segment from a plurality of path segments included in each second current individual; generating a second target path segment based on two path segment nodes corresponding to the first target path segment; In a case where the second target path segment does not intersect with the grids where the at least one obstacle object is located, the first target path segment included in each second current individual is replaced by the second target path segment to obtain the third current individual.
2. The method according to claim 1, It is characterized in that The generating a plurality of initial paths from the initial grid to the target grid in the confined space by taking the range of the confined space and the respective positions of the at least one obstacle object as constraints comprises: Using the target grid as an attraction source, and using the grid where the inner wall of the enclosed space is located and the grid where the at least one obstacle object is located as repulsion sources, to obtain a potential field in the enclosed space; In the process of generating each initial path, using an artificial potential field method, a plurality of intermediate grids are determined from the grid system based on the potential field; The initial path is obtained based on the initial grid, the target grid and the plurality of intermediate grids.
3. The method according to claim 2, It is characterized in that The method of using an artificial potential field method to determine a plurality of intermediate grids from the grid system based on the potential field comprises: For the i-th intermediate grid, the potential field stress of the i-th intermediate grid in the potential field is determined by using the artificial potential field method, wherein i is a positive integer; Determine the transfer probability from the i-th middle grid to each of the multiple adjacent grids based on the potential field force and the movement direction of the i-th middle grid to each of the multiple adjacent grids, wherein the multiple adjacent grids are all adjacent to the i-th middle grid; Based on the transition probabilities from the i-th intermediate grid to each of the plurality of adjacent grids, an i+1-th intermediate grid is selected from the plurality of adjacent grids.
4. The method according to claim 1, It is characterized in that The step of performing a crossover operation on the plurality of first current individuals based on the respective fitness of the plurality of first current individuals to obtain a plurality of second current individuals comprises: Based on the multiple first current individuals, determining multiple individual groups, wherein the individual group consists of two first current individuals; Determining the selection probability of each of the plurality of individual groups based on the fitness of each of the plurality of first current individuals; determining a plurality of target individual groups from the plurality of individual groups based on the respective selection probabilities of the plurality of individual groups; A crossover operation is performed on two first current individuals respectively included in the plurality of target individual groups to obtain the plurality of second current individuals.
5. The method according to claim 4, It is characterized in that The performing crossover operations on two first current individuals respectively included in the plurality of target individual groups to obtain the plurality of second current individuals comprises: For each target individual group, when the same grid exists in two paths corresponding to the two first current individuals included in the individual group, a cross operation is performed on the paths corresponding to the two first current individuals included in the individual group based on the same grid to obtain two second current individuals.
6. A motion control method, It is characterized in that The motion control method comprises: In response to a motion control request, obtaining a target path from an initial grid where the robot is located in a grid system to a target grid where a target position is located in the grid system; Using a fuzzy algorithm, controlling the robot to move along the target path to the target position; The target path is obtained by using the path determination method according to any one of claims 1 to 5.
7. The method according to claim 6, It is characterized in that The target path includes a plurality of path segments formed based on a plurality of path nodes; Wherein, the using of fuzzy algorithm to control the robot to move along the target path to the target position includes: When the robot is controlled to move to the kth path node, obstacle information of the kth path segment is acquired by the robot, wherein the k path segments represent a path between the kth path node and the k+1th path node in the target path, wherein k is a positive integer; When the obstacle information of the k-th path segment indicates that there is an obstacle in the k-th path segment, a fuzzy algorithm is used to perform local path planning based on the k-th path node and the k+1-th path node to obtain a target path; The robot is controlled to move along the target path to the k+1th path node.
