Robot path planning method based on combination of particle swarm, ant colony and A-Star algorithm

Abstract

The application belongs to the technical field of path planning, and discloses a robot path planning method based on combination of particle swarm, ant colony and A-Star algorithm, which comprises the following steps: generating a grid map, obtaining positions of obstacles and feasible areas in the grid map, constructing an accessible distance matrix, determining starting positions and target positions of each robot in the grid map in a robot cluster; and using the particle swarm, ant colony combination algorithm and A-Star algorithm to perform path optimization planning on the robot cluster based on the accessible distance matrix, the starting positions and the target positions of the robots in the grid map, so as to obtain an optimal path. The application combines the ant colony algorithm, the algorithm and the simulated annealing algorithm with the particle swarm algorithm, and performs robot obstacle avoidance path optimization, so that the robot can quickly find a path from the starting point to the end point while completing obstacle avoidance.

Description

Technical Field

[0001] This invention belongs to the field of path planning technology, and in particular relates to a robot path planning method based on a combination of particle swarm optimization, ant colony optimization and A-Star algorithm. Background Technology

[0002] Currently, robots can be broadly defined as mechanical devices that can automatically complete tasks according to requirements. Robots can assist or replace people in the work they do in daily life. In recent years, with the rapid development of cloud computing, big data, the Internet of Things, and artificial intelligence, intelligentization is considered an important direction for the development of robotics technology today.

[0003] One of the most common problems in robot design is path planning. Path planning is a major problem in robotics research. The goal is to find the lowest-cost, conflict-free path from a known starting point, avoiding all obstacles, to reach a known destination. Traditional methods for solving robot path planning problems generally include Dijkstra's algorithm (a breadth-first search algorithm that searches a large area to find the shortest path from the target point to all reachable points; it's a greedy search algorithm, and its efficiency drops significantly when the search area is large because it may find invalid data for robot movement), artificial potential field method (when the robot is at a certain position, if the resultant repulsive force of several obstacles is equal in magnitude and opposite in direction to the gravitational force of the target point, then the resultant force is 0. According to the algorithm rules, the robot will no longer be "forced," preventing it from continuing to search for obstacle avoidance paths), A* algorithm (which causes significant performance overhead when the target point is unreachable), and ant colony algorithm (the heuristic function is usually related to the distance of the path, typically the reciprocal of the distance. The disadvantage of this is that it can cause premature convergence, easily leading to local optima; if this heuristic function is used in robot path planning problems, it can easily cause the robot to enter a deadlock state), etc.

[0004] Based on the above analysis, the problems and defects of the existing technology are as follows: the existing path planning method has a long planning time, low efficiency, and is prone to causing the robot to enter a deadlock state, making it impossible to obtain path planning results. Summary of the Invention

[0005] To address the problems existing in the prior art, this invention provides a robot path planning method based on a combination of particle swarm optimization, ant colony optimization, and A-Star algorithms.

[0006] This invention is implemented as follows: a robot path planning method based on a combination of particle swarm optimization, ant colony optimization, and A-Star algorithms, wherein the robot path planning method based on the combination of particle swarm optimization, ant colony optimization, and A-Star algorithms includes:

[0007] Generate a grid map and obtain the positions of obstacles and feasible areas in the grid map, construct a reachability distance matrix, and determine the starting position and target position of each robot in the robot cluster in the grid map;

[0008] By combining particle swarm optimization, ant colony optimization, and A-Star algorithm, the optimal path is obtained for the robot swarm based on the reachability distance matrix, the robot's starting position in the grid map, and the target position.

[0009] Furthermore, the robot path planning method based on a combination of particle swarm optimization, ant colony optimization, and A-Star algorithms includes the following steps:

[0010] Step 1: Generate a grid map and obtain the positions of obstacles and feasible areas in the grid map. Number the generated grid map. Establish a two-dimensional coordinate system with the lower left corner of the generated grid map as the origin (converting the actual problem into a graphical model for easier calculation).

[0011] Step 2: Initialize the parameters of the particle swarm optimization algorithm and the ant colony optimization algorithm based on the grid map; calculate the reachability matrix based on the acquired grid map (the data is pre-processed to facilitate calculation during algorithm execution);

[0012] Step 3: On the grid map, the combined particle swarm optimization, ant colony optimization, and A-Star algorithm are used to perform path optimization planning for the robot cluster based on the reachability distance matrix, the robot's starting position and target position on the grid map, and obtain the optimal path (the three algorithms are reasonably bridged and improved with our own improved formula, which greatly improves the accuracy and fast convergence of the algorithm).

[0013] Furthermore, the parameters for initializing the particle swarm optimization and ant colony optimization algorithms based on the grid map include:

[0014] Initialize the parameters of the particle swarm optimization and ant colony optimization algorithms based on the order of the raster map;

[0015] The parameters of the particle swarm include: the size of the particle swarm, the maximum value of the individual acceleration factor, the minimum value of the individual acceleration factor, the maximum value of the swarm acceleration factor, the minimum value of the swarm acceleration factor, and the maximum number of iterations of the algorithm.

[0016] The parameters of the ant colony include: colony size, acceleration factor, pheromone concentration importance factor, heuristic function importance factor, inertia coefficient, pheromone volatility factor, pheromone concentration, and maximum number of iterations.

[0017] Furthermore, the method of using a combination of particle swarm optimization, ant colony optimization, and A-Star algorithm on a grid map to perform path optimization planning for the robot swarm based on the reachability distance matrix, the robot's starting position, and the target position on the grid map includes:

[0018] (1) Place the particles at the starting point of the grid map and control each particle to move according to the given direction of movement;

[0019] (2) Calculate the transition probability of each reachable point of the particle, select the next target point to visit based on the transition probability, and put the target point into the visited taboo list. Determine whether all particles have reached the target point or whether all particles are feasible points. If so, execute step (3); otherwise, return to step (1) until all particles have reached the target point.

[0020] (3) Calculate the estimated minimum number of moves based on the coordinates of the starting point and the target point, calculate the path length and number of moves for each particle, and record the best and worst solutions in the current ant colony;

[0021] (4) Calculate the corresponding inertia coefficient of each particle, and determine whether to copy the path corresponding to the current best particle based on the inertia coefficient of each particle;

[0022] (5) Update the pheromone concentration of adjacent reachable points:

[0023] (6) Determine whether the maximum number of iterations or the termination condition has been reached. If not, return to step (2). Otherwise, terminate the iteration and output the path planning result.

[0024] Furthermore, the method of using a combination of particle swarm optimization, ant colony optimization, and A-Star algorithm on a grid map to perform path optimization planning for the robot swarm based on the reachability distance matrix, the robot's starting position, and the target position on the grid map includes the following steps:

[0025] 1) Place each particle at the starting point of the grid, calculate the transition probability of the particle to each reachable point using the following formula, calculate the next target point visited by the particle using the roulette wheel algorithm, and add the target point to the visited taboo list:

[0026] ;

[0027] in, express Moment Particle From grid Transfer to grid The probability of; Represents particles The set of grid cells to be visited after removing obstacle grid points; Factors indicating the importance of pheromone concentration; A factor representing the importance of a heuristic function; express Momentary particles from the grid Transfer to grid An improved heuristic function, , Indicates the current grid To feasible grid The distance; Represents a feasible grid Euclidean distance to the target point; Indicates the individual acceleration factor. ; Indicates the population acceleration factor. ; , , and These represent the pre-set acceleration factors. and The maximum and minimum values; Indicates the current iteration number. This represents the maximum number of iterations given in advance;

[0028] 2) Repeat step 1) until all particles have reached the target point or the current particle is a feasible point, then proceed to step 3).

[0029] 3) Calculate the estimated minimum number of moves based on the coordinates of the starting point and the target point. :

[0030] ;

[0031] in, Indicates the x-coordinate number of the starting point. Indicates the ordinate number of the starting point; Indicates the x-coordinate index of the target point. Indicates the ordinate number of the target point;

[0032] 4) Calculate the path length of each particle. and number of moves Record the best and worst solutions in the current ant colony:

[0033] ;

[0034] in, Represents particles The number of moves from the starting point to the target point. Represents particles The current number reached is One grid, Indicates from grid Move to grid The distance;

[0035] 5) Decide whether to copy the current best particle based on its own inertia coefficient;

[0036] 6) Update pheromone concentration using the following formula:

[0037] ;

[0038] in, ( () represents an important factor in pheromone concentration, used to characterize the degree of pheromone volatility. Indicates the first Sub-iteration grid To grid The concentration of pheromones Indicates the first A particle in the grid With grid The concentration of pheromones released along the connection path The calculation formula is as follows:

[0039] ;

[0040] in, This indicates the estimated minimum number of moves. This represents the pheromone concentration constant. Represents particles The number of times the particle moves from the starting point to the target point If the target point has not been reached, then ;

[0041] 7) Determine if the maximum number of iterations or the termination condition has been reached. If so, terminate the iteration and output the path planning result; otherwise, return to step 1.

[0042] Furthermore, the step of determining whether to replicate the current optimal particle based on its own inertia coefficient includes:

[0043] 5.1) Calculate the inertia coefficient corresponding to the current particle. :

[0044] ;

[0045] in, This represents the maximum value of a pre-given coefficient of inertia. This represents the minimum value of a pre-defined coefficient of inertia. Represents particles The total length of the path from the starting point to the target point. This represents the maximum length of the path found by all ants so far.

[0046] 5.2) Generate a random number. =rand(0, 1);

[0047] 5.3) Determine the inertia coefficient corresponding to the current particle. The relationship between the size of the generated random numbers and the value of the random numbers, if > If the current particle copies the path taken by the best particle in the current particle swarm, then the current particle will copy the path taken by the best particle in the current particle swarm; otherwise, the current result will be accepted.

[0048] Another objective of this invention is to provide a robot path planning system based on the combination of particle swarm optimization, ant colony optimization, and A-Star algorithm for implementing the aforementioned robot path planning method. The robot path planning system based on the combination of particle swarm optimization, ant colony optimization, and A-Star algorithm includes:

[0049] The grid map construction and processing module is used to generate grid maps and obtain the positions of obstacles and feasible areas in the grid maps, number the generated grid maps, and establish a two-dimensional coordinate system with the lower left corner of the generated grid map as the origin of the coordinate system.

[0050] The initialization module is used to initialize the parameters of the particle swarm optimization and ant colony optimization algorithms based on the grid map.

[0051] The reachability matrix construction module is used to calculate the reachability matrix based on the acquired raster map;

[0052] The path planning module is used to optimize the path planning of the robot cluster on the grid map by using a combination of particle swarm optimization, ant colony optimization and A-Star algorithm based on the reachability distance matrix, the robot's starting position and target position on the grid map, and obtain the optimal path.

[0053] The output module is used to output the optimal planned path.

[0054] Another object of the present invention is to provide a computer device including a memory and a processor, the memory storing a computer program, which, when executed by the processor, causes the processor to perform the steps of the robot path planning method based on a combination of particle swarm optimization, ant colony optimization, and A-Star algorithms.

[0055] Another object of the present invention is to provide a computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the steps of the robot path planning method based on a combination of particle swarm optimization, ant colony optimization, and A-Star algorithms.

[0056] Another objective of this invention is to provide an information data processing terminal for implementing the robot path planning system based on a combination of particle swarm optimization, ant colony optimization, and A-Star algorithms.

[0057] Based on the above technical solutions and the technical problems solved, the advantages and positive effects of the technical solution to be protected by this invention are as follows:

[0058] First, addressing the technical problems existing in the prior art and the difficulty in solving them, this paper closely analyzes, in conjunction with the technical solution to be protected by this invention and the results and data obtained during the research and development process, how the technical solution of this invention solves the technical problems, and the inventive technical effects brought about by solving these problems. The specific description is as follows:

[0059] This invention uses a combination of particle swarm optimization and ant colony optimization algorithms to perform path optimization planning for robot swarms, resulting in an obstacle avoidance scheme after K iterations of the particle swarm optimization algorithm. This ensures that the robot avoids obstacles while finding the optimal path. This invention can improve the working efficiency of robots, reduce unnecessary losses during movement, and save time and resources.

[0060] Meanwhile, this invention has a certain degree of dynamism, and can dynamically modify the position of the target point by modifying the function coordinates of the endpoint in the estimated minimum number of iterations, thereby increasing the number of iterations of the algorithm, affecting the acceleration factor, and achieving the effect of dynamic path optimization.

[0061] This invention also introduces tabu vectors to prevent the robot from moving onto obstacles. The feasible region is dynamically calculated during the computation process, ensuring the robot does not move onto obstacles.

[0062] Second, considering the technical solution as a whole or from a product perspective, the technical effects and advantages of the technical solution to be protected by this invention are specifically described as follows:

[0063] This invention combines the principles of ant colony optimization, Pythagorean theorem, and simulated annealing with particle swarm optimization to optimize obstacle avoidance paths for robots. This allows the robot to quickly find a path from the starting point to the destination while simultaneously avoiding obstacles. By modifying the heuristic function, pheromone concentration update strategy, and acceleration factor update strategy, this invention increases the robot's attraction to the target point, resulting in faster algorithm convergence.

[0064] Third, as supplementary evidence of the inventive step of the claims of this invention, it is also reflected in the following important aspects:

[0065] (1) The expected benefits and commercial value of the technical solution of this invention after transformation are as follows:

[0066] This invention is applied to the field of 2D path optimization for robots. It pre-models the environment as a 2D Boolean grid matrix, requiring minimal storage space and thus saving storage costs. After specifying the start and end points, the program quickly calculates an unobstructed path, reducing time costs.

[0067] (2) The technical solution of the present invention solves a technical problem that people have long wanted to solve but have never been able to solve successfully:

[0068] To a certain extent, it solves the problems of particle swarm optimization (PSO) being difficult to solve discrete problems (in combination with ant colony and A* algorithms) and having fast convergence speed but easy to enter local optima (in combination with simulated annealing algorithm). Attached Figure Description

[0069] Figure 1 This is a schematic diagram of the robot path planning method based on the combination of particle swarm optimization, ant colony optimization, and A-Star algorithm provided in the embodiments of the present invention.

[0070] Figure 2 This is a flowchart of a robot path planning method based on a combination of particle swarm optimization, ant colony optimization, and A-Star algorithm provided in an embodiment of the present invention.

[0071] Figure 3 It is a 4-4 numbered grid matrix provided in the embodiments of the present invention;

[0072] Figure 4 This is a schematic diagram of the robot's directional movement in a grid map provided in an embodiment of the present invention;

[0073] Figure 5 This is a walking route map planned for a robot in a given 20th-order grid matrix, provided by an embodiment of the present invention.

[0074] Figure 6 This is a walking route map planned for a robot in a given 40th order grid matrix, provided by an embodiment of the present invention;

[0075] Figure 7 This is a convergence curve of the algorithm provided in an embodiment of the present invention in a given 20th order raster matrix;

[0076] Figure 8 This is a convergence curve of the algorithm provided in an embodiment of the present invention in a given 40th order raster matrix;

[0077] Figure 9 This is a flowchart illustrating the process of using the particle swarm optimization algorithm combined with simulated annealing to solve for the minimum value of a function, as provided in this embodiment of the invention.

[0078] Figures 10-13 This is a convergence trajectory diagram of the particle swarm optimization algorithm combined with simulated annealing algorithm provided in this embodiment of the invention for solving the minimum value of four given test functions. Detailed Implementation

[0079] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0080] I. Explanatory and Illustrative Embodiments. To enable those skilled in the art to fully understand how the present invention is specifically implemented, this section provides an explanatory and illustrative description of the embodiments described in the claims.

[0081] like Figures 1-2 As shown, the robot path planning method based on a combination of particle swarm optimization, ant colony optimization, and A-Star algorithms provided in this embodiment of the invention includes the following steps:

[0082] S101, Generate a grid map and obtain the positions of obstacles and feasible areas in the grid map, and number the generated grid map; establish a two-dimensional coordinate system with the lower left corner of the generated grid map as the origin of the coordinate system;

[0083] S102, Initialize the parameters of the particle swarm optimization algorithm and the ant colony optimization algorithm based on the grid map; Calculate the reachability matrix based on the acquired grid map;

[0084] S103 uses a combination of particle swarm optimization, ant colony optimization, and A-Star algorithm on a grid map to perform path optimization planning for the robot cluster based on the reachability distance matrix, the robot's starting position and target position on the grid map, and obtains the optimal path.

[0085] The robot path planning method based on a combination of particle swarm optimization, ant colony optimization, and A-Star algorithms provided in this embodiment of the invention specifically includes the following steps:

[0086] Step 1: Set up the system simulation environment. First, set the order of the raster matrix. The generated raster matrix is ​​of size . The variables in the grid matrix are Boolean variables, where 0 represents a feasible grid and 1 represents an obstacle. Next, the grid map and the positions of obstacles and feasible regions within it are obtained, along with the robot's starting and target positions. A grid map (a 4x4 grid map) is generated based on the obstacle and feasible region positions. Figure 3As shown), based on the location of the obstacle and the location of the feasible area; the rule for numbering the assumptions is to start from 1 from left to right and from bottom to top, and to establish a two-dimensional coordinate system with the bottom left corner as the origin, thereby obtaining the coordinate transformation formula. A two-dimensional coordinate system is established for the generated grid matrix according to the following formula, assuming the numbering rule is to start from 1 from left to right and from bottom to top, and to establish a two-dimensional coordinate system with the bottom left corner as the origin:

[0087] ;

[0088] in The grid number. This represents the side length of each grid cell, typically taken as... .

[0089] Step 2: Set the parameters required for the particle swarm optimization (PSO) and ant colony optimization (ACO) algorithms based on the order of the raster map; set the required parameters for the algorithms:

[0090] Among them, the parameters of the particle swarm include: the size of the particle swarm. The maximum value of the individual acceleration factor Minimum value of individual acceleration factor The maximum value of the population acceleration factor Minimum value of the population acceleration factor Maximum number of iterations of the algorithm .

[0091] Ant colony parameters include: colony size. Accelerator and Important factors of pheromone concentration Heuristic function importance factor Inertia coefficient and pheromone volatile factors pheromone concentration Maximum number of iterations wait.

[0092] Step 3: Calculate the reachability matrix based on the grid map. Assume the robot follows... Figure 4 The method follows an octree for single-step optimization, meaning the robot can move to adjacent grids in eight directions during the search process, as shown in the figure.

[0093] Assuming the robot can only move one step at a time, then the maximum distance it can move is... and If grid Unreachable grid Then set the distance between them to Calculate the values ​​in the raster matrix separately. Each grid cell (excluding obstacles) is at a distance from other The distance of grid cells yields a size of ... Symmetric reachable matrix That is, when the grid size is: When the size of the raster matrix is Raster maps, such as Figure 3 As shown.

[0094] Table 1 shows the distances from grid 1 to other grids.

[0095]

[0096] The distance from grid 1 to other grids (N represents unreachable).

[0097] Step 4: Apply a path optimization algorithm to the grid map:

[0098] Step 4 can be divided into the following sub-steps:

[0099] 4.1 For each particle Generate a path vector (vector size set to) Used to store particles The grid cells visited in this cycle, from the starting point to the ending point. For each particle... Generate a tabu list vector (vector size is) The taboo vector is used to store particles. The set of grid cells already visited in this loop. Generate the pheromone concentration matrix. The size of the matrix is And the value of each element in the matrix is ​​equal and greater than 0 (the pheromone concentration is equal on the initial path). Generate path vectors (vector size set to) This is used to store the optimal path length found by the particle swarm in each iteration. A path of length is generated. path length vector Each element in the vector stores the first... The minimum path length found by the particle swarm optimization in the next iteration. Store the optimal path found after the algorithm ends and the length of the optimal path, respectively.

[0100] 4.2 Place each particle at the starting point of the grid (set the value of the first element of each particle's path table as the index of the starting grid point), and select the next grid to visit by calculating the pheromone concentration between each grid. Initially, the pheromone concentration on the connecting paths between all grids is the same, i.e. ,use Represents grid With grid The concentration of pheromones along the connection path.

[0101] 4.3 Calculate the transition probability of each particle to each reachable point, and use the roulette wheel algorithm to calculate the next move point of the particle, until all particles reach the target point, including:

[0102] The formula for calculating the transition probability to a feasible grid point is:

[0103]

[0104] in, express Moment Particle From grid Transfer to grid The probability of; for Momentary particles from the grid Transfer to grid An improved heuristic function; Represents particles The set of grid cells to be visited (excluding obstacle grid cells), over time The elements in the grid will gradually decrease until they reach the target grid point. This represents the particle. The target grid has been reached. The factor representing the importance of pheromone concentration is as follows: the higher the value, the greater the role of pheromone concentration in particle transfer. This is a factor indicating the importance of the heuristic function; the larger the value, the greater the role of the heuristic function in the transition.

[0105] The calculation formula is as follows:

[0106] ;

[0107] in, Indicates the individual acceleration factor; Indicates the population acceleration factor. For the current grid To feasible grid The distance is determined by the raster matrix. To determine, For feasible grids Euclidean distance to the target point Represents particles The current set of movable grid cells. This improvement prevents greedy strategies from causing particles to easily get stuck in a "deadlock" state during the initial optimization process.

[0108] set up Vector represents the current particle The state transition probability vector. The current particle The sum of probabilities of transitioning to each feasible grid point, let ,according to The value of the particle is used to determine the particle. Which grid point should be selected next, and what should be the selected grid point? Add to path vector In the middle, the grid Add tabu list vector middle.

[0109] , The calculation formula is as follows:

[0110]

[0111] .

[0112] in , , and These are the pre-defined acceleration factors. and The maximum and minimum values. This is the current iteration number. This is the maximum number of iterations given in advance. The above formula can dynamically adjust the speedup factor as the number of iterations of the algorithm increases. and Perform corrections.

[0113] The above-described scheme in this embodiment of the invention enables the particles to focus on their global search capabilities at the beginning of the algorithm due to the randomness of their positions. This emphasizes the particle's search range and enhances the search performance of each particle, preventing premature convergence and avoidance of the algorithm quickly falling into local optima. Therefore, as the number of iterations increases, the influence of the position of the optimal particle in the particle swarm on the search direction of each particle should be increased. Based on the above analysis, It should be gradually reduced and It should be increased gradually.

[0114] Repeat the above process until the particles... If the current target grid is found or there is no feasible grid, it is determined whether the current particle has visited the target grid or whether the current particle is already a feasible point. If so, proceed to step 4.4; otherwise, continue the loop.

[0115] 4.4 Calculate the estimated minimum number of moves based on the coordinates of the starting point and the target point. :

[0116]

[0117] in, The x-coordinate number of the starting point. The ordinate number is the starting point's ordinate. The x-coordinate number of the target point. The ordinate number is the ordinate number of the target point.

[0118] 4.5 Calculate the path length of each particle and number of moves Record the best and worst solutions in the current ant colony:

[0119] For each particle in the particle swarm Repeat the above process until every particle in the swarm has completed path optimization and updated. , , .

[0120] Particle path length Refers to particles The total length of the path from the starting point to the target point. Representative particles The number of moves from the starting point to the target point is determined by the following formula:

[0121]

[0122] in Representative particles The current number reached is 1 grid (assuming the numbering rule is from left to right, bottom to top, starting from 1), For from grid Move to grid The distance, in the above formula for .

[0123] 4.6 Determine whether to replicate the current optimal particle based on its own inertia coefficient:

[0124] Calculate the inertia coefficient corresponding to the current particle. Drawing inspiration from the simulated annealing algorithm, a random number is generated. =rand(0, 1), if > If the current solution is rejected, the particle is made to copy the path taken by the best particle in the current particle swarm. Otherwise, the current solution is accepted.

[0125] particle Corresponding inertia coefficient Determined by the following formula:

[0126]

[0127] in, This represents the maximum value of a pre-given coefficient of inertia. This represents the minimum value of a pre-defined coefficient of inertia. Refers to particles The total length of the path from the starting point to the target point. This refers to the maximum length of the path found by all ants so far.

[0128] The advantage of the above technical solution adopted in the embodiments of the present invention is that each particle Each corresponds to a coefficient of inertia. When particles When the found path length is small, the corresponding A larger ant is more likely to maintain its chosen path, while a smaller ant is less likely to maintain its chosen path. When the found path length is large, the corresponding The smaller size allows the algorithm to expand its search range and converge faster as the number of iterations increases.

[0129] 4.7 Calculating Particles The length of the path traveled by the particle, if If the target point is not reached this time, the path length is set to [value]. ;

[0130] 4.8 For each particle in the particle swarm Repeat the above process until every particle in the swarm has completed path optimization and updated. , , .

[0131] Step 5, update pheromone concentration:

[0132] ;

[0133] in ( () is an important factor in pheromone concentration, representing the degree of pheromone volatility. For the first Sub-iteration grid To grid The concentration of pheromones Indicates the first A particle in the grid With grid The concentration of pheromones released along the connection path is determined by the following formula:

[0134]

[0135] in, To estimate the minimum number of moves, The pheromone concentration constant is... For particles The number of times the particle moves from the starting point to the target point, if the particle If the target point has not been reached, then .

[0136] The advantage of the above-described scheme in the embodiments of the present invention is that when the heuristic function has a significant factor When the value is reasonable and the current particle Choosing a better or optimal path can greatly increase the probability that the ant will choose that path in the next few iterations, and this operation also greatly increases the convergence speed of the algorithm.

[0137] Step 6: Determine if the algorithm has reached the maximum number of iterations or the termination condition. If not, return to step 4; otherwise, terminate the iteration and output the result.

[0138] II. Application Examples. To demonstrate the inventiveness and technical value of the technical solution of this invention, this section provides application examples of the technical solution of the claims on specific products or related technologies.

[0139] The robot path planning method based on a combination of particle swarm optimization, ant colony optimization, and A-Star algorithms provided in this invention was applied to function optimization and tested using Matlab software. It demonstrated strong superiority on commonly used two-dimensional test functions. Furthermore, the algorithm was tested using Matlab software, obtaining optimal path planning schemes on both 20th and 40th order raster matrices (verified using the Dijkstra algorithm).

[0140] This invention applies a robot path planning method based on a combination of particle swarm optimization, ant colony optimization, and A-Star algorithms to a computer device. The computer device includes a memory and a processor. The memory stores a computer program. When the computer program is executed by the processor, the processor executes the robot path planning method based on a combination of particle swarm optimization, ant colony optimization, and A-Star algorithms.

[0141] This invention applies a robot path planning method based on a combination of particle swarm optimization, ant colony optimization, and A-Star algorithms to a computer-readable storage medium containing a computer program. When the computer program is executed by a processor, the processor executes the robot path planning method based on the combination of particle swarm optimization, ant colony optimization, and A-Star algorithms.

[0142] This invention applies a robot path planning method based on a combination of particle swarm optimization, ant colony optimization, and A-Star algorithm to an information data processing terminal.

[0143] It should be noted that embodiments of the present invention can be implemented in hardware, software, or a combination of both. The hardware portion can be implemented using dedicated logic; the software portion can be stored in memory and executed by a suitable instruction execution system, such as a microprocessor or dedicated-design hardware. Those skilled in the art will understand that the above-described devices and methods can be implemented using computer-executable instructions and / or included in processor control code, for example, such code provided on a carrier medium such as a disk, CD, or DVD-ROM, a programmable memory such as read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The devices and modules of the present invention can be implemented by hardware circuitry such as very large-scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, or programmable hardware devices such as field-programmable gate arrays, programmable logic devices, etc., or by software executed by various types of processors, or by a combination of the above-described hardware circuitry and software, such as firmware.

[0144] III. Evidence of the Relevant Effects of the Embodiments. The embodiments of the present invention have achieved some positive effects during research and development or use, and indeed possess significant advantages compared to existing technologies. The following description, in conjunction with data, charts, and other materials from the experimental process, illustrates these advantages.

[0145] The robot path planning method based on a combination of particle swarm optimization, ant colony optimization, and A* algorithm provided in this invention was tested using Matlab software. It demonstrated strong superiority on commonly used two-dimensional test functions. Furthermore, the algorithm was tested in Matlab software, achieving optimal path planning schemes on both 20th and 40th order grid matrices (verified using the Dijkstra algorithm). Compared to other traditional path planning algorithms, it possesses a certain degree of dynamic optimization (if the endpoint position needs to be changed, only the minimum number of moves required to move the starting and target points needs to be modified in the algorithm). The x and y coordinates of the midpoint are used to increase the number of iterations in the algorithm. Thus, the acceleration factor can be modified. This achieves the effect of dynamic optimization of the algorithm. During the calculation process, the feasible region is dynamically calculated, preventing the robot from moving to obstacles.

[0146] One 20th-order grid matrix and one 40th-order grid matrix are randomly generated. For each grid matrix, the robot's initial point and target point are set. The algorithm parameters are set as follows: number of particles. =70, pheromone importance factor =1, heuristic function importance factor =7, the initial values ​​of the pheromone concentration matrix are all set to 8, and the pheromone volatility coefficient is... =0.2, pheromone concentration =1, =0.4, , =0.215, =1.5, =0.75, =2.5, maximum number of iterations =100.

[0147] Simulation content and result analysis:

[0148] The particle swarm optimization algorithm is used for robot path planning. The algorithm plans the robot's path as follows: Figure 5 and Figure 6 As shown in the figure, black grids represent obstacles, white grids represent feasible grids, red solid lines represent the robot's path, purple solid dots represent the starting point, and green solid dots represent the ending point. The two results in the figure above were verified using the Dijkstra algorithm, showing that the robot found the optimal path in both instances. Figure 7 and Figure 8 The convergence trajectories of the algorithm for solving 20th and 40th order raster matrices are given. As can be seen from the figure, the algorithm has high convergence.

[0149] The improved particle swarm optimization algorithm provided in this embodiment of the invention provides the following parameters for solving the problem:

[0150] , , , , , Maximum speed Take 10% of the difference in each dimension of the search range of the function to be tested, and the initial temperature. , , .

[0151] Sphere functions:

[0152]

[0153] Schaffer function:

[0154]

[0155] Griewank function:

[0156]

[0157] Rastrigrin function:

[0158]

[0159] The above four functions are solved using the algorithm provided in the embodiments of the present invention, and compared with the GA, PSO, LDWPSO and S-PSO algorithms.

[0160] Simulation analysis:

[0161] like Figures 10 to 13 As shown in Table 2, the improved particle swarm optimization algorithm (SAPSO) provided in this embodiment of the invention has lower average, variance, and standard deviation than the GA, PSO, LDWPSO, and S-PSO algorithms in the optimization processes of the Sphere, Schaffer, and Griewank functions. It has been mentioned that the simulated annealing-based particle swarm optimization algorithm (SAPSO) has high solution accuracy. However, SAPSO's optimization performance on the Rastrigin function is inferior to the other four algorithms. Analysis reveals that although the optimal solution obtained by the SAPSO algorithm in 30 tests was very close to 0 in 16 of them, it failed to find the optimal solution in the other 14 tests. This indicates that the stability of SAPSO needs further improvement.

[0162] Table 2. Statistics on solving the minimum values ​​of four given test functions using the Particle Swarm Optimization (PSO) algorithm combined with Simulated Annealing (SAM) algorithm.

[0163]

[0164] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications, equivalent substitutions, and improvements made by those skilled in the art within the scope of the technology disclosed in the present invention, and within the spirit and principles of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A robot path planning method based on a combination of particle swarm optimization, ant colony optimization, and A-Star algorithm, characterized in that, The robot path planning method based on a combination of particle swarm optimization, ant colony optimization, and A-Star algorithms includes: Generate a grid map and obtain the positions of obstacles and feasible areas in the grid map, construct a reachability distance matrix, and determine the starting position and target position of each robot in the robot cluster in the grid map; By combining particle swarm optimization and ant colony optimization algorithms with the A-Star algorithm, the optimal path is obtained by optimizing the path of the robot swarm based on the reachability distance matrix, the robot's starting position and target position in the grid map. The robot path planning method based on a combination of particle swarm optimization, ant colony optimization, and A-Star algorithms includes the following steps: Step 1: Generate a grid map and obtain the positions of obstacles and feasible areas in the grid map. Number the generated grid map. Establish a two-dimensional coordinate system with the lower left corner of the generated grid map as the origin. Step 2: Initialize the parameters of the particle swarm optimization algorithm and the ant colony optimization algorithm based on the grid map; calculate the reachability matrix based on the acquired grid map; Step 3: On the grid map, the combined particle swarm optimization and ant colony optimization algorithms and the A-Star algorithm are used to optimize the path of the robot cluster based on the reachability distance matrix, the starting position of the robot in the grid map and the target position, so as to obtain the optimal path. The method of using a combination of particle swarm optimization, ant colony optimization, and A-Star algorithms on a grid map to perform path optimization planning for a robot swarm based on the reachability distance matrix, the robot's starting position, and the target position on the grid map includes: (1) Place the particles at the starting point of the grid map and control each particle to move according to the given direction of movement; (2) Calculate the transition probability of each reachable point of the particle, select the next target point to visit based on the transition probability, and put the target point into the visited taboo list; determine whether all particles have reached the target point or whether all particles are feasible points. If so, execute step (3); otherwise, return to step (1) until all particles have reached the target point. (3) Calculate the estimated minimum number of moves based on the coordinates of the starting point and the target point, calculate the path length and number of moves for each particle, and record the best and worst solutions in the current ant colony; (4) Calculate the corresponding inertia coefficient of each particle, and determine whether to copy the path corresponding to the current best particle based on the inertia coefficient of each particle; (5) Update the pheromone concentration of adjacent reachable points: (6) Determine whether the maximum number of iterations or the termination condition has been reached. If not, return to step (2). Otherwise, terminate the iteration and output the path planning result. Whether to copy the current best particle based on its own inertia coefficient includes: Calculate the inertia coefficient corresponding to the current particle. : ； in, This represents the maximum value of a pre-given coefficient of inertia. This represents the minimum value of a pre-defined coefficient of inertia. Represents particles The total length of the path from the starting point to the target point. This represents the maximum length of the path found by all ants so far. Generate a random number =rand(0, 1); Determine the inertia coefficient corresponding to the current particle. The relationship between the size of the generated random numbers and the value of the random numbers, if > If the current particle copies the path taken by the best particle in the current particle swarm, then the current particle will copy the path taken by the best particle in the current particle swarm; otherwise, the current result will be accepted.

2. The robot path planning method based on a combination of particle swarm optimization, ant colony optimization, and A-Star algorithm as described in claim 1, characterized in that, The parameters for initializing the particle swarm optimization and ant colony optimization algorithms based on the grid map include: Initialize the parameters of the particle swarm optimization and ant colony optimization algorithms based on the order of the raster map; The parameters of the particle swarm include: the size of the particle swarm, the maximum value of the individual acceleration factor, the minimum value of the individual acceleration factor, the maximum value of the swarm acceleration factor, the minimum value of the swarm acceleration factor, and the maximum number of iterations of the algorithm. The parameters of the ant colony include: colony size, acceleration factor, pheromone concentration importance factor, heuristic function importance factor, inertia coefficient, pheromone volatility factor, pheromone concentration, and maximum number of iterations.

3. The robot path planning method based on a combination of particle swarm optimization, ant colony optimization, and A-Star algorithm as described in claim 1, characterized in that, The method of using a combination of particle swarm optimization, ant colony optimization, and A-Star algorithm on a grid map to perform path optimization planning for a robot swarm based on the reachability distance matrix, the robot's starting position, and the target position on the grid map includes the following steps: 1) Place each particle at the starting point of the grid, calculate the transition probability of the particle to each reachable point using the following formula, calculate the next target point visited by the particle using the roulette wheel algorithm, and add the target point to the visited taboo list: ； in, express Moment Particle From grid Transfer to grid The probability of; Represents particles The set of grid cells to be visited after removing obstacle grid points; Factors indicating the importance of pheromone concentration; A factor representing the importance of a heuristic function; express Momentary particles from the grid Transfer to grid An improved heuristic function , Indicates the current grid To feasible grid The distance; Represents a feasible grid Euclidean distance to the target point; Indicates the individual acceleration factor. ; Indicates the population acceleration factor. ; , , and These represent the pre-set acceleration factors. and The maximum and minimum values; Indicates the current iteration number. This represents the maximum number of iterations given in advance; 2) Repeat step 1) until all particles have reached the target point or the current particle is a feasible point, then proceed to step 3). 3) Calculate the estimated minimum number of moves based on the coordinates of the starting point and the target point. : ； in, Indicates the x-coordinate number of the starting point. Indicates the ordinate number of the starting point; Indicates the x-coordinate index of the target point. Indicates the ordinate number of the target point; 4) Calculate the path length of each particle. and number of moves Record the best and worst solutions in the current ant colony: ； in, Represents particles The number of moves from the starting point to the target point, Represents particles The current number reached is One grid, Indicates from grid Move to grid The distance; 5) Decide whether to copy the current best particle based on its own inertia coefficient; 6) Update pheromone concentration using the following formula: ； in, ( () represents an important factor in pheromone concentration, used to characterize the degree of pheromone volatility. Indicates the first Sub-iteration grid To grid The concentration of pheromones Indicates the first A particle in the grid With grid The concentration of pheromones released along the connection path The calculation formula is as follows: ； in, This indicates the estimated minimum number of moves. This represents the pheromone concentration constant. Represents particles The number of times the particle moves from the starting point to the target point If the target point has not been reached, then ; 7) Determine if the maximum number of iterations or the termination condition has been reached. If so, terminate the iteration and output the path planning result; otherwise, return to step 1.

4. A robot path planning system based on a combination of particle swarm optimization, ant colony optimization, and A-Star algorithm, implementing the robot path planning method based on a combination of particle swarm optimization, ant colony optimization, and A-Star algorithm as described in claim 1, characterized in that, The robot path planning system based on a combination of particle swarm optimization, ant colony optimization, and A-Star algorithms includes: The grid map construction and processing module is used to generate grid maps and obtain the positions of obstacles and feasible areas in the grid maps, number the generated grid maps, and establish a two-dimensional coordinate system with the lower left corner of the generated grid map as the origin of the coordinate system. The initialization module is used to initialize the parameters of the particle swarm optimization and ant colony optimization algorithms based on the grid map. The reachability matrix construction module is used to calculate the reachability matrix based on the acquired raster map; The path planning module is used to optimize the path planning of the robot cluster on the grid map by using a combination of particle swarm optimization, ant colony optimization and A-Star algorithm based on the reachability distance matrix, the robot's starting position and target position on the grid map, and obtain the optimal path. The output module is used to output the optimal planned path.

5. A computer device, characterized in that, The computer device includes a memory and a processor. The memory stores a computer program, which, when executed by the processor, causes the processor to perform the steps of the robot path planning method based on the combination of particle swarm, ant colony, and A-Star algorithms as described in any one of claims 1 to 3.

6. A computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the steps of the robot path planning method based on a combination of particle swarm optimization, ant colony optimization, and A-Star algorithms as described in any one of claims 1 to 3.

7. An information data processing terminal, characterized in that, The information data processing terminal is used to implement the robot path planning system based on the combination of particle swarm, ant colony and A-Star algorithms as described in claim 4.

Images