Path planning method based on improved ant colony algorithm

Abstract

The invention relates to the technical field of path planning, in particular to a path planning method based on an improved ant colony algorithm, which comprises the following steps: establishing a grid map, and initializing parameters of the ant colony algorithm; determining an initial value of pheromone concentration according to the distance between the nodes in the grid map and the initial node and the distance between the nodes in the grid map and the target node; ants are placed at the initial node for multiple times, each ant starts from the initial node, the next node where the ant moves is selected according to the pheromone concentration and heuristic information, and the taboo table and the current pheromone concentration are updated; emptying the taboo table after the tracking of all ants is completed, and recording the paths of all ants; updating a pheromone volatilization factor based on a proportional relation between the average path length and the minimum path length; after the maximum number of iterations is reached, a final planned path is obtained according to the recorded path through which the ants walk; the convergence speed and reliability of the ant colony algorithm can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of path planning, and particularly relates to a path planning method based on an improved ant colony algorithm. Background Art

[0002] In the fields of robots, autonomous vehicles, and logistics transportation, path planning is an important technology. It aims to find the optimal or near-optimal path from the starting point to the ending point to meet various constraints, such as the shortest distance, the least time, or the lowest cost, etc. The ant colony algorithm was proposed by Marco Dorigo in 1992 and can be used to solve path planning problems. The ant colony algorithm is a heuristic search algorithm that simulates the foraging behavior of ants. It realizes global optimization by simulating the mechanism that ants release pheromones during the process of finding food and use this to guide other ants to find the shortest path.

[0003] However, there are some deficiencies in the application of the traditional ant colony algorithm to path planning. In the traditional ant colony algorithm, the initial value of pheromone is evenly distributed. Due to the same initial pheromone concentration, the search randomness is strong and relatively blind in the initial stage of the algorithm, resulting in a significant decrease in the convergence speed and optimization ability of the algorithm. In addition, as a constant parameter, the setting of the pheromone evaporation factor has a significant impact on the performance of the algorithm. If it is set too large, it will cause the pheromone to evaporate too fast, weakening the guiding effect of excellent paths and affecting the convergence speed of the algorithm; if it is set too small, it will cause the pheromone to evaporate too slowly, resulting in excessive accumulation of path pheromones in the previous iteration process and easily falling into local optimality. Summary of the Invention

[0004] In view of the above problems, the present invention provides a path planning method based on an improved ant colony algorithm, which solves the technical problems of slow convergence speed and easy falling into local optimality of the ant colony algorithm in the prior art.

[0005] The present invention provides a path planning method based on an improved ant colony algorithm, including:

[0006] Step S1, establish a grid map of the warehouse. The grid map includes multiple nodes, and determine the types of nodes according to the warehouse terrain corresponding to the nodes. The types of nodes include initial nodes, target nodes, obstacle nodes, and non-obstacle nodes; initialize the parameters of the ant colony algorithm;

[0007] Step S2, determine the initial value of pheromone concentration according to the distances between the nodes in the grid map and the initial node and the target node;

[0008] The initial value of the pheromone concentration is determined by the initial pheromone distribution function, and the expression is:

[0009]

[0010] Among them, τ i represents the initial pheromone concentration on node i; τ0 represents the initial pheromone concentration of node i in the ant colony algorithm, and the initial pheromone concentration is the same for each node; represents the pheromone concentration increased according to the distance relationship between node i and the starting point O; represents the pheromone concentration increased according to the distance relationship between node i and the target node D; (x i , y i ), (x O , y O ), (x D , y D ) represent the coordinates of node i, the initial node O, and the target node D respectively;

[0011] Step S3: Place ants at the initial node multiple times to perform the tracing from the initial node to the target node. The tracing process includes: Each ant starts from the initial node, selects the next node for the ant to move according to the current pheromone concentration and heuristic information, and updates the taboo list; Update the current pheromone concentration based on the current pheromone evaporation factor;

[0012] After completing the tracing of all ants, clear the taboo list and record the paths traveled by all ants;

[0013] Step S4: Obtain the contemporary average path length and the contemporary minimum path length based on the path lengths traveled by all ants, and update the pheromone evaporation factor according to the proportional relationship between the contemporary average path length and the contemporary minimum path length;

[0014] Step S5: Return to step S3 until the maximum number of iterations is reached. Take the optimal path in the last iteration convergence process as the finally planned path, and use the finally planned path to drive the warehouse robot to complete navigation.

[0015] Preferably, step S1 specifically includes:

[0016] Step S1-1: Divide the warehouse floor area into square grids of uniform size, and each grid serves as a node; Determine the grid area where the fixed obstacles in the warehouse are located as obstacle nodes, determine the entrances and exits of the warehouse as the initial nodes, determine the picking location as the target node, and determine the area where the robot in the warehouse can move freely as non-obstacle nodes;

[0017] Step S1-2: Initialize the parameters of the ant colony algorithm, including: Initialize the number of ants m, the information heuristic factor α and the heuristic function factor β of the state transition formula, the pheromone intensity Q, and the maximum number of iterations NC, and assign initial values.

[0018] Preferably, step S3 specifically includes:

[0019] Step S3-1: Place virtual ants at the initial node. Each ant determines the state transition probability based on the current pheromone concentration and heuristic information, selects the next node according to the state transition probability, updates the taboo list based on all the nodes that the ant has passed through so far, and after tracing all the nodes, updates the current pheromone concentration based on the current pheromone evaporation factor;

[0020] Step S3-2: Return to step S3-1 until the taboo list is cleared after the ants have traced a preset number of times, and record the paths that all the ants have passed through.

[0021] Preferably, in step S3-1, the expression for the state transition probability is:

[0022]

[0023] where is the probability that the k-th ant selects node j as the next position from node i at time t, τ i,j (t) represents the pheromone concentration on the path (i, j) at time t, η i,j represents the heuristic function for selecting node j as the next position from node i, α is the information heuristic factor, and β is the heuristic function factor; allowed k ={0, 1, 2,..., n - 1}-tabu k represents the set of nodes that ant k is allowed to select next, tabu k ,k = 1, 2,..., m represents the taboo list, where k is the index of the current ant and m is the total number of ants, and n is the total number of nodes.

[0024] Preferably, in step S3-1, the expression for updating the current pheromone concentration based on the current pheromone evaporation factor is:

[0025] τ i,j (t + 1)=(1 - ρ)·τ i,j (t)+Δτ i,j (t)

[0026] where τ i,j (t + 1) represents the pheromone concentration on the path (i, j) at time t + 1, ρ represents the pheromone evaporation factor; 1 - ρ represents the pheromone residue coefficient; Δτ i,j (t) represents the total amount of pheromone left by all the ants on the path (i, j) at time t.

[0027] Preferably, the calculation expression for the total amount of pheromone Δτ i,j (t) left by all the ants on the path (i, j) at time t is:

[0028]

[0029] Among them, represents the amount of pheromone left by ant k on path (i, j), and its calculation expression is:

[0030]

[0031] Among them, Q represents the total amount of pheromone released by each ant after walking a complete path; L k represents the total length of the path walked by the k-th ant in this pheromone trail search.

[0032] Preferably, step S4 specifically includes:

[0033] Step S4-1: Calculate the average value and the minimum value of the path lengths walked by all the ants, and respectively obtain the contemporary average path length and the contemporary minimum path length;

[0034] Step S4-2: Use the proportional relationship between the contemporary average path length and the contemporary minimum path length as the path concentration degree; update the pheromone evaporation factor according to the magnitude relationship between the path concentration degree and the preset critical concentration degree.

[0035] Preferably, in step S4-2, the expression for updating the pheromone evaporation factor is:

[0036]

[0037] ε = l ave / l min

[0038] Among them, ρ e is the pheromone evaporation factor in the early stage; ρ a is the pheromone evaporation factor in the later stage; ε is the path concentration degree, l ave is the average path length of the contemporary search; l min is the minimum path length of the contemporary search, and ε0 is the preset critical concentration degree value.

[0039] Compared with the prior art, the present invention has at least the following beneficial effects:

[0040] (1) The present invention dynamically adjusts the initial pheromone concentration based on the distances from nodes to the starting point and the ending point. Nodes closer to the starting and ending points are assigned higher initial pheromone concentrations, reducing the probability of distal nodes being selected in the initial stage of the algorithm and enhancing the attraction of nodes on the better paths near the starting and ending points to ants. This strategy enables ants to focus more quickly on more potential areas, enhancing the directivity of the better paths in the initial stage of the ant colony algorithm, effectively reducing the blindness of ants when selecting paths in the initial stage of the algorithm, and enabling the ant colony to find potential high-quality paths more quickly.

[0041] (2) The present invention introduces a dual evaporation factor strategy, which dynamically adjusts the size of the evaporation factor according to the concentration degree of each generation of paths. In the early stage of iteration, a higher evaporation factor accelerates the evaporation of pheromones on each path, encouraging the ant colony to conduct a more extensive global search and avoiding premature convergence to local optimal solutions. In the later stage of iteration, a lower evaporation factor is conducive to the accumulation of pheromones on the better paths, thus retaining the information of the better paths and avoiding local optima.

[0042] (3) The present invention optimizes the initial pheromone concentration, enhances the guidance of the search, and uses the dual evaporation factor strategy to dynamically balance the global search and local optimization. The present invention integrates these improvements into the existing ant colony algorithm, improving the efficiency of the ant colony algorithm, the quality of the solution, and enhancing the performance and robustness of the algorithm when solving complex optimization problems, enabling it to find the optimal solution faster and more accurately. Description of the Drawings

[0043] The drawings are only for the purpose of illustrating specific embodiments and are not considered to be a limitation of the present invention.

[0044] Figure 1 It is a flowchart of the path planning method based on the improved ant colony algorithm provided by the present invention.

[0045] Figure 2 It is a detailed flowchart of the path planning method based on the improved ant colony algorithm provided by the present invention.

[0046] Figure 3 It is a schematic diagram of the grid map provided by the present invention.

[0047] Figure 4 It is a schematic diagram of the path planning result of the traditional ant colony algorithm.

[0048] Figure 5 It is a schematic diagram of the path planning result of the improved algorithm provided by the present invention.

[0049] Figure 6 It is a schematic diagram of the change trend of the convergence curves of the traditional ant colony algorithm and the improved algorithm of the present invention provided by the present invention. Detailed Embodiments

[0050] In order to more clearly understand the above objects, features, and advantages of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, without conflict, the embodiments of the present invention and the features in the embodiments can be combined with each other. In addition, the present invention can also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited by the specific embodiments disclosed below.

[0051] Aiming at the disadvantages of the ant colony algorithm in the convergence process, such as strong randomness and easy to fall into local optimum, the present invention provides a path planning method based on an improved ant colony algorithm. The initial pheromone concentration parameter of the ant colony algorithm is adjusted according to the distances from each node to the starting point and the target point to reduce the blindness in the initial stage of the algorithm. A double evaporation factor strategy is proposed to adjust the pheromone evaporation factor of the algorithm in two stages divided according to the path concentration degree, so that the sizes of the parameters match optimally to improve the algorithm convergence speed and path planning accuracy.

[0052] As Figure 1 、 Figure 2 shown, the present invention provides a path planning method based on an improved ant colony algorithm, and the specific implementation steps are as follows:

[0053] Step S1: Establish a grid map and initialize the parameters of the ant colony algorithm; the grid map includes multiple nodes, and the types of the nodes include an initial node, a target node, an obstacle node, and a non-obstacle node.

[0054] The grid map established by the present invention is as Figure 3 shown. The grid map is a square grid structure, and each grid is used as a node. The grid map has an initial node and a target node. The initial node is the starting point of the path, and the target node is the destination of the path. The grid map also has obstacle nodes, which are points that cannot be used as paths, and non-obstacle nodes, which are nodes other than obstacle nodes and can be used as paths.

[0055] In some embodiments, the present invention first establishes a square grid map, designates an initial node and a target node in the grid map, and then the obstacle nodes can be determined by means of random generation or the like.

[0056] In some embodiments, the path planning of the present invention is for the path planning of a warehouse robot. The warehouse environment is rasterized, that is, the warehouse floor area is divided into square grids of uniform size. The size of the grid can be determined according to the size of the robot, the navigation accuracy requirements, and the complexity of the warehouse space. Then, according to the positions of the actual objects in the warehouse, the type of each grid node is determined. For example, the grid areas where fixed obstacles such as shelves, walls, and columns in the warehouse are located can be marked as obstacle nodes, and the robot cannot pass through. The entrances and exits of the warehouse can be set as initial nodes, while the locations where goods are stored or picked up can be set as target nodes. The areas in the warehouse where the robot can freely pass, except for those occupied by obstacles and the initial and target nodes, are divided into non-obstacle nodes. Finally, the physical space information of the warehouse is converted into a discrete raster map that can be understood and processed by a computer.

[0057] After establishing the raster map, the present invention initializes multiple parameters of the ant colony algorithm. For other basic parameters except for the pheromone concentration and the pheromone evaporation factor, including the number of ants m, the information heuristic factor α and the heuristic function factor β of the state transition formula, the pheromone intensity Q, and the maximum number of iterations NC, they are all initialized and given initial values.

[0058] Step S2: Determine the initial value of the pheromone concentration according to the distance between the nodes in the raster map and the initial node and the target node.

[0059] In the traditional ant colony algorithm, the initial pheromone concentration is uniformly distributed, and the search randomness is strong and relatively blind in the initial stage of the algorithm. To reduce the blindness of the search in the initial stage of the algorithm and enhance the directivity of the pheromone to the ants, the present invention provides an initial pheromone distribution function, and the expression is:

[0060]

[0061]

[0062] Among them, τ i represents the initial pheromone concentration on node i; τ0 represents the initial pheromone concentration of node i in the ant colony algorithm, and the initial pheromone concentration is the same for each node; represents the increased pheromone concentration according to the distance relationship between node i and the starting point O; represents the increased pheromone concentration according to the distance relationship between node i and the target node D; (x i , y i ), (x O , y O ), (x D , y D ) represent the coordinates of node i, the initial node O, and the target node D respectively.

[0063] In the above way, the initial value of the pheromone concentration of each node in the present invention is affected by the distances from it to the starting point and the ending point. For nodes closer to the starting and ending points, the initial value of the pheromone concentration is larger. By improving the initial pheromone concentration, the probability of remote nodes being selected in the initial stage of iteration is reduced, and the attraction of nodes on the relatively optimal path in the initial stage to ants is enhanced.

[0064] Step S3: Place virtual ants at the initial node multiple times to perform the tracing from the initial node to the target node. The tracing process includes: Each ant starts from the initial node, selects the next node for the ant to move according to the current pheromone concentration and heuristic information, and updates the taboo list; update the current pheromone concentration based on the current pheromone evaporation factor; after all ants complete the tracing, clear the taboo list and record the paths passed by all ants.

[0065] In this step, virtual ants are placed at the initial node of the grid map multiple times and traced. The specific steps are described as follows.

[0066] Each ant selects the next node according to the pheromone and heuristic information through the state transition probability formula, and records all the nodes passed by each ant currently to form the taboo list tabu k , k = 1, 2,..., m, where k is the index of the current ant and m is the total number of ants. The taboo list records all the nodes passed by each ant currently; the expression of the state transition probability is:

[0067]

[0068] Among them, is the probability that the k-th ant selects node j as the next position from node i at time t, τ i,j (t) represents the pheromone concentration on the path (i, j) at t, η i,j represents the heuristic function for selecting node j as the next position from node i. α is the information heuristic factor, which reflects the role of the pheromone accumulated by the ant in the tracing process in choosing the path. The larger α is, the more the ant tends to choose the path with a larger pheromone concentration; β is the heuristic function factor, which represents the role of the heuristic information in path selection. The larger β is, the greater the influence of the path length on the selection probability; allowed k ={0, 1, 2,..., n - 1}-tabu k represents the set of nodes that ant k is allowed to select next, and n is the total number of nodes.

[0069] After each ant completes the traversal of all nodes, update the residual pheromone. The pheromone τ on the path (i, j) at time t + 1 i,j(t + 1) is updated as follows:

[0070] τ i,j (t + 1) = (1 - ρ)·τ i,j (t) + Δτ i,j (t)

[0071] where ρ represents the pheromone evaporation factor, which is a variable coefficient with a value range of (0, 1); 1 - ρ represents the pheromone residue coefficient; Δτ i,j (t) represents the total amount of pheromone left by all ants on path (i, j) at time t. Δτ i,j The calculation expression of Δτ

[0072]

[0073] where represents the amount of pheromone left by ant k on path (i, j). The calculation expression of

[0074]

[0075] where Q represents the total amount of pheromone released by each ant after walking a complete path, which is a constant and will affect the convergence speed of the algorithm to a certain extent; L k represents the total length of the path taken by the k-th ant in this search. If the total length of the path selected by this ant is larger, then the amount of pheromone it releases per unit length of the route is smaller.

[0076] After all the ant searches are completed, the taboo list tabu k is reset, the content of the taboo list is cleared, and the path lengths walked by all ants are recorded.

[0077] Step S4: Obtain the contemporary average path length and the contemporary minimum path length based on the path lengths of all the paths walked by the ants, and update the pheromone evaporation factor according to the proportional relationship between the contemporary average path length and the contemporary minimum path length.

[0078] In this step, after one iteration of all the ant searches is completed, the present invention updates the pheromone evaporation factor.

[0079] The evaporation factor ρ of the traditional ant colony algorithm is a fixed value, which is not conducive to global search in the early stage of iteration and has a slow convergence speed in the later stage. This is because the evaporation factor ρ of the ant colony algorithm is related to the convergence speed and global search ability. When the value of ρ is large, the pheromone evaporates too fast, reducing the pheromone content on the better paths, resulting in the possible exclusion of the optimal path. When the value of ρ is small, the pheromone content on each path is not much different, and the algorithm is likely to fall into local optimum, affecting the quality of the solution.

[0080] Based on this, the present invention provides a dual evaporation factor strategy. The iteration process is divided into two stages, the early stage and the later stage, according to the concentration degree of each generation of paths. In the early stage of iteration, the evaporation factor is large, accelerating the evaporation of pheromone on each path, facilitating the ant colony to carry out global search; in the later stage, the evaporation factor is small, facilitating the accumulation of pheromone, thus accelerating convergence and improving the quality of the solution. The improved pheromone evaporation factor is:

[0081]

[0082] ε = l ave / l min

[0083] where ρ e is the evaporation factor of pheromone in the early stage; ρ a is the evaporation factor of pheromone in the later stage; ε is the path concentration degree, l ave is the average path length of the current search; l min is the minimum path length of the current search, and ε0 is a preset critical concentration value.

[0084] Step S5: Return to step S3 until the maximum number of iterations is reached. Take the optimal path in the convergence process of the last iteration as the finally planned path, and use the finally planned path to drive the warehouse robot to complete navigation.

[0085] The ant colony iteration process of the present invention completes iteration update after reaching the maximum number of iterations.

[0086] In some embodiments, the optimal path searched by the ants in the last generation is used as the finally planned path.

[0087] Embodiment 1

[0088] In this embodiment, the present invention provides a path planning method based on an improved ant colony algorithm applied to the path planning of a warehouse robot, and the specific implementation manner is described as follows.

[0089] Step S100: Construct a grid map and initialize the parameters of the ant colony algorithm.

[0090] In this step, a 1m x 1m square grid is overlaid on the warehouse floor plan, and each 1m x 1m grid square is defined as a node. The areas occupied by fixed shelves, walls, and support columns are marked as obstacle nodes. For example, if a shelf occupies grid cells (2,2), (2,3), (3,2), (3,3), these nodes are set as obstacle nodes. The typical charging station for the robot or the designated loading and unloading area in the warehouse is set as the initial node, and the specific locations on the shelves where the robot needs to pick up or unload goods are set as target nodes. All other grid squares in the warehouse that are not marked as obstacle, initial, or target nodes are set as non-obstacle nodes, which are areas where the robot can navigate.

[0091] Initialize the parameters of the ant colony algorithm. Set the number of ants m = 50, the information heuristic factor α of the state transition formula = 1, the heuristic function factor β = 2, the pheromone intensity Q = 100, and the maximum number of iterations NC = 100.

[0092] Step S200: Determine the initial pheromone concentration based on the distances to the starting point and the target node.

[0093] Assign (x,y) coordinates to each node in the grid. For example, (1,1) is the node in the upper left corner, (1,2) is the node to its right, and so on. Then calculate the initial pheromone concentration based on the coordinates of each node. The calculation formula has been given before.

[0094] Step S300: Conduct multiple ant path searches.

[0095] Place virtual ants at the initial node multiple times to trace the path from the initial node to the target node. The tracing process includes: Each ant starts from the initial node, selects the next node for the ant to move to based on the current pheromone concentration and heuristic information, and updates the taboo list; Update the current pheromone concentration based on the current pheromone evaporation factor; After all ants have completed tracing, clear the taboo list and record the paths taken by all ants. The specific scheme and calculation formula for this step have been given before.

[0096] Step S400: Update the pheromone evaporation factor according to the concentration degree of the paths.

[0097] Set the critical concentration value ε0 = 1.2. Based on the formula given before, determine the early pheromone evaporation factor and the late pheromone evaporation factor based on the critical concentration value, and update the final pheromone evaporation factor.

[0098] Step S500: Complete the iteration and determine the optimal path.

[0099] Return to step S300, and repeat ant search, pheromone update, and evaporation factor adjustment for a total of NC iterations. Eventually, the optimal path is obtained, and the optimal path can be stored in the form of a series of ordered grid node coordinates, such as [(1,1),(1,2),(2,2),...,(15,20)].

[0100] Step S600: Drive the robot to complete the warehouse freight navigation based on the optimal path.

[0101] In this step, parse the path data, extract the continuous node sequence that constitutes the path, and convert the grid coordinates into actual coordinates in the physical environment of the warehouse. For example, each 1m x 1m grid cell corresponds to the actual size in the warehouse. According to the recorded grid node coordinates (such as (1,1)), determine the absolute position in the physical space, and finally generate a sequence of physical path points.

[0102] According to the sequence of physical path points and combined with the kinematic model and control method of the robot, generate specific control commands that the robot can execute, such as speed control commands, position control commands, steering control commands, etc. Send the generated robot control commands to the robot to finally complete the warehouse freight navigation.

[0103] The robot motion trajectory obtained through experiments in the present invention is as Figure 4 , Figure 5 shown, Figure 4 is the path planning result of the traditional ant colony algorithm, Figure 5 is the path planning result of the improved algorithm provided by the present invention. Figure 6 is the change trend of the convergence curves of the traditional ant colony algorithm and the improved algorithm provided by the present invention. Table 1 shows the comparison of the results of the traditional ant colony algorithm and the improved algorithm of the present invention.

[0104] Table 1

[0105]

[0106] Although the specific implementation manners of the present invention depict each action or step in a specific order, it should be understood that such actions or steps are required to be executed in the specific order shown or in a sequential order, or all the illustrated actions or steps should be executed to obtain the desired result. In certain environments, multitasking and parallel processing may be beneficial. Similarly, although several specific implementation details are included in the above discussion, these should not be construed as limiting the scope of the present disclosure. Certain features described in the context of separate embodiments can also be implemented in combination in a single implementation. Conversely, the various features described in the context of a single implementation can also be implemented separately or in any suitable sub-combination in multiple implementations.

[0107] As described above, it is only the preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any change or replacement that can be easily thought of by those skilled in the art within the technical scope disclosed by the present invention should be covered within the protection scope of the present invention.

Claims

1. A path planning method based on an improved ant colony algorithm, characterized in that: The following steps are involved: Step S1, establishing a grid map of the warehouse, the grid map includes a plurality of nodes, determining the type of the node according to the warehouse terrain corresponding to the node, the node type includes an initial node, a target node, an obstacle node and a non-obstruction node; initializing the parameters of the ant colony algorithm; Step S2, determining an initial value of pheromone concentration according to the distances between nodes other than the initial node and the target node in the grid map and the initial node and the target node; The initial value of the pheromone concentration is determined by the initial pheromone distribution function, which is expressed as: Among them, τ i represents the initial pheromone concentration on node i; τ0 represents the initial pheromone concentration of node i in the ant colony algorithm, and the initial pheromone concentration is the same for each node; represents the pheromone concentration increased according to the distance relationship between node i and the initial node O; represents the pheromone concentration increased according to the distance relationship between node i and target node D; (x i ,y i ),(x O ,y O ),(x D ,y D ) represent the coordinates of node i, initial node O and target node D respectively; Step S3, placing ants at the initial node multiple times to search from the initial node to the target node, wherein the search process includes: each ant starts from the initial node, selects the next node to which the ant moves according to the current pheromone concentration and heuristic information, and updates the taboo table; updates the current pheromone concentration based on the current pheromone volatility factor; After all ants have completed their tracing, the taboo table is cleared to record the paths that all ants have taken; Step S4, obtaining a contemporary average path length and a contemporary minimum path length based on the path lengths traveled by all the ants, and updating the pheromone volatilization factor according to the proportional relationship between the contemporary average path length and the contemporary minimum path length; Step S5, return to step S3, until the maximum number of iterations is reached, take the optimal path of the last iterative convergence process as the final planned path, and use the final planned path to drive the warehouse robot to complete navigation.

2. The path planning method based on the improved ant colony algorithm according to claim 1, characterized in that: Step S1 specifically includes: Step S1-1, divide the warehouse plane area into square grids of uniform size, each grid as a node; determine the grid area where the fixed obstacles in the warehouse are located as the obstacle node, determine the entrance and exit of the warehouse as the initial node, determine the pickup location as the target node, and determine the area in the warehouse where the robot can pass freely as the non-obstruction node; Step S1-2, initializing the parameters of the ant colony algorithm, including: initializing the number of ants m, the information heuristic factor α and the heuristic function factor β of the state transition formula, the pheromone strength Q and the maximum number of iterations NC, and assigning initial values.

3. The path planning method based on the improved ant colony algorithm according to claim 2 is characterized in that: Step S3 specifically includes: Step S3-1, placing virtual ants at the initial node, each ant determines the state transition probability based on the current pheromone concentration and heuristic information, selects the next node according to the state transition probability, updates the taboo table according to all the nodes currently passed by the ant, and after completing the tracking of all nodes, updates the current pheromone concentration based on the current pheromone volatility factor; Step S3-2, return to step S3-1, until the ants complete the preset number of traces, then clear the taboo table and record the paths taken by all the ants.

4. The path planning method based on the improved ant colony algorithm according to claim 3 is characterized in that: In step S3-1, the expression of the state transition probability is: in, is the probability that the kth ant chooses node j as the next position from node i at time t, τ i,j (t) represents the pheromone concentration of t on the path (i, j), η i,j represents the heuristic function for node i to select node j as the next position, α is the information heuristic factor, β is the heuristic function factor; allowed k ={0,1,2,...,n-1}-tabu k represents the set of nodes that ant k is allowed to choose next, tabu k ,k=1,2,...,m represents the taboo table, where k is the index of the current ant, m is the total number of ants, and n is the total number of nodes.

5. The path planning method based on the improved ant colony algorithm according to claim 4 is characterized in that: In step S3-1, the expression for updating the current pheromone concentration based on the current pheromone volatilization factor is: t i,j (t+1)=(1-ρ)·τ i,j (t)+Δτ i,j (t) Among them, τ i,j (t+1) represents the pheromone concentration on the path (i, j) at time t+1, ρ represents the pheromone volatility factor; 1-ρ represents the pheromone residual coefficient; Δτ i,j (t) represents the total amount of pheromone left by all ants on the path (i, j) at time t.

6. The path planning method based on the improved ant colony algorithm according to claim 5 is characterized in that: The total amount of pheromone Δτ left by all ants on the path (i, j) at time t i,j The calculation expression of (t) is: in, represents the amount of pheromone left by ant k on path (i, j), The calculation expression is: Where Q represents the total amount of pheromone released by each ant after completing a complete path; L k It represents the total length of the path taken by the kth ant in this search.

7. The path planning method based on the improved ant colony algorithm according to claim 6, characterized in that: Step S4 specifically includes: Step S4-1, calculating the average value and the minimum value of the path lengths walked by all the ants, and obtaining the contemporary average path length and the contemporary minimum path length respectively; Step S4-2, taking the proportional relationship between the contemporary average path length and the contemporary minimum path length as the path concentration; and updating the pheromone volatilization factor according to the relationship between the path concentration and the preset critical concentration.

8. The path planning method based on the improved ant colony algorithm according to claim 7, characterized in that: In step S4-2, the expression for updating the pheromone volatility factor is: e=l ave / l min Among them, ρ e is the early pheromone volatilization factor; a is the late pheromone volatilization factor; ε is the path concentration, l ave is the average path length of contemporary search; l min is the minimum path length of contemporary search, and ε0 is the preset critical concentration value.

Images