Unmanned driving path planning method based on cluster collaborative ant colony algorithm

By introducing adaptive step size, adaptive ants count and bidirectional ant colony collaboration strategy in unmanned driving path planning, combining node distance index and probability expansion, the fixed step size, insufficient exploration and slow iteration in traditional ant colony algorithm are solved. The problem of this problem achieves more efficient path planning and global optimal solution.

CN119002472APending Publication Date: 2024-11-22ANHUI NORMAL UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410902477.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-07
Publication Date
2024-11-22

AI Technical Summary

Technical Problem

Traditional ant colony algorithm has a fixed step-length limit flexibility in unmanned driving path planning, resulting in low complexity and efficiency of path planning. Inadequate exploration in the early stage leads to local optimal solutions, slow iteration convergence speed and easy to fall into local optimal solutions. The problem of excessive pheromones concentration cannot be identified.

Method used

Introduce adaptive step size strategy, adaptive ant number strategy and two-way ant colony collaboration strategy, use node distance index instead of pheromone to build heuristic functions, and dynamically adjust the number and iterative strategy, combined with probability expansion optimization processing, to achieve faster Convergent and global optimal solutions.

Benefits of technology

It significantly improves the smoothness and efficiency of path planning, avoids local optimal solutions, improves the convergence speed and stability of the algorithm, shortens the operation time, and achieves more efficient path planning.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure BDA0004930950130000023
    Figure BDA0004930950130000023
  • Figure BDA0004930950130000024
    Figure BDA0004930950130000024
  • Figure BDA0004930950130000026
    Figure BDA0004930950130000026
Patent Text Reader

Abstract

The invention discloses an unmanned driving path planning method based on a cluster collaborative ant colony algorithm. The method aims at solving the problems that an ant colony algorithm is low in convergence speed, poor in ant flexibility, prone to falling into a local optimal solution and the like in unmanned driving path planning. The invention provides a cluster collaborative ant colony algorithm for unmanned driving global path planning. An adaptive step length strategy is adopted to enable an ant colony to get rid of fixed step length limitation; the ant colonies share node information through a two-way ant colony cooperation strategy, so that advancing has guidance; an adaptive ant colony number strategy is adopted to increase the number of ant colony exploration in the initial stage of iteration and reduce the number of ant colonies in the later stage of iteration; and reconstructing a heuristic function dominated by the node distance index. A simulation experiment shows that the algorithm not only can realize global rapid convergence, but also has high stability and shorter operation time, and the quality of an obtained solution and the convergence speed of the algorithm are more superior than those of a traditional ant colony algorithm.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of unmanned driving autonomous navigation technology, and specifically to an unmanned driving path planning method based on a cluster collaborative ant colony algorithm. Background Art

[0002] As a key technology for autonomous driving, path planning provides a basis for autonomous driving to complete driving tasks. This technology aims to find the optimal path from the initial position to the target position in the configuration space through a planning algorithm without colliding with any obstacles. The design of the algorithm is the core of path planning. Common path planning algorithms include ant colony algorithm, genetic algorithm, Dijkstra algorithm, A* algorithm, artificial potential field method and particle swarm algorithm.

[0003] The ant colony algorithm is a natural swarm intelligence algorithm that simulates the behavior of ants. During the foraging process, ants release a chemical called pheromone along the path. The shorter the path, the more pheromones are released. Finally, the uneven distribution of pheromones will gradually attract more and more ants to move to shorter paths, and finally find the optimal path. The ant colony algorithm has strong applicability and robustness, and has a wide range of applications in the study of global path planning problems.

[0004] However, the ant colony algorithm has the following shortcomings in the problem of unmanned driving path planning:

[0005] (1) In the traditional ant colony algorithm framework, the moving step of ants is fixed to 1, which means that each time a node is transferred, except for the previous node, its activity range is limited to the seven adjacent grids around it. This setting greatly restricts the flexibility of ants, not only increasing the overall complexity of the algorithm, but also seriously affecting the smoothness and efficiency of path planning.

[0006] (2) In the traditional ant colony algorithm, since the number of ant colonies in each iterative exploration is the same, there is a situation where the map is not completely traversed in the early exploration of the algorithm, and the results of the later exploration are relatively concentrated, which will lead to local optimal solutions and waste of some computing resources.

[0007] (3) The traditional ant colony algorithm has a slow iterative convergence speed and is prone to falling into local optimal solutions. In addition, the algorithm is blind during the exploration process, which greatly reduces the convergence speed of the algorithm.

[0008] (4) In the ant colony algorithm, when the pheromone concentration is too high, it may have a negative effect. Even if the algorithm is faced with a better solution, it will not be able to correctly identify it due to the high pheromone content on the current path, and thus cannot obtain the global optimal solution. Summary of the invention

[0009] In order to solve the above-mentioned shortcomings of the traditional ant colony algorithm, the present invention proposes an unmanned driving path planning method based on cluster collaborative ant colony algorithm.

[0010] In view of the fact that the moving step length of ants in the traditional ant colony algorithm is fixed to 1, which restricts the flexibility of ants, increases the overall complexity of the algorithm, and affects the smoothness and efficiency of path planning, the present invention introduces an adaptive step length strategy and changes the node selection logic of the ant colony algorithm to select nodes that can be reached in a straight line from the current node and are not blocked by obstacles as optional nodes for the next movement. The improved comparison diagram is shown in the attached figure. Figure 1 This method enables the present invention to avoid the selection of redundant nodes, reduce path turning points and the optimal path length.

[0011] In view of the problem that the traditional ant colony algorithm does not completely traverse the map in the early exploration of the algorithm, and the results of the later exploration are relatively concentrated, resulting in the situation of local optimal solutions. The present invention introduces an adaptive step size strategy and an adaptive ant colony quantity strategy, which increases the number of ant colonies in the early exploration and reduces the number of ant colonies in the later exploration by dynamically adjusting the number of ant colonies in the iterative process. This method enables the present invention to search extensively in the early exploration stage to obtain more feasible solutions, and avoid a large number of identical convergence results in the later exploration stage, which leads to the waste of some computing resources.

[0012] In view of the problems of blindness, slow iterative convergence and easy to fall into local optimal solutions in the exploration process of traditional ant colony algorithms, the present invention introduces a bidirectional ant colony collaboration strategy, divides the ant colony into forward and reverse dual populations, alternately performs iterative exploration, and introduces the use of node distance index. This method enables the present invention to have clear guidance during bidirectional iterative exploration, and quickly converge to the optimal solution in alternating iterations.

[0013] Finally, in view of the problem that the traditional ant colony algorithm cannot obtain the global optimal solution due to the excessively high pheromone content in the local area, the present invention proposes and uses the node distance index of the bidirectional ant colony to replace the pheromone to construct a new heuristic function, which is composed of the node distance index and the Euclidean distance of the current node from the node to be transferred, and introduces the probability extension of the optimization process. This method enables the cluster collaborative ant colony algorithm of the present invention to approach the optimal solution faster, which can greatly improve the efficiency of the optimization algorithm and accelerate the convergence speed.

[0014] The technical solution adopted by the present invention is an unmanned driving path planning method based on cluster collaborative ant colony algorithm, comprising the following steps:

[0015] Step 1: Receive the path planning request of the driverless vehicle and obtain map-related information through the Internet of Vehicles function.

[0016] Step 2: Create a grid map and obtain a multi-step adjacency matrix based on the adaptive step strategy.

[0017] Step 3: Use the traditional ant colony algorithm to perform preliminary path planning, with the aim of updating the node distance index data matrix of the forward and reverse populations and providing a data basis for the cluster collaborative ant colony algorithm.

[0018] Step 4: During the iteration of the cluster collaborative ant colony algorithm, the number of ants currently exploring is adjusted according to the adaptive ant number strategy. This method enables the algorithm to reduce computing resources and speed up convergence.

[0019] Step 5: Adopt a two-way ant colony collaboration strategy, divide the ant colony into forward and reverse populations, and perform node transfer according to the heuristic function dominated by the node distance index. After the node is transferred, record the node it has moved to and update the node distance index data matrix.

[0020] Step 6: Reconstruct the heuristic function ε composed of the node distance index and the Euclidean distance between the current node and the node to be transferred ij , and introduces the probability extension of optimization processing. This method makes the ant node transfer more clearly oriented, allowing the algorithm to converge to the optimal solution faster.

[0021] Furthermore, the map-related information obtained includes information such as the starting point, the end point, the road, the fork road nodes and the obstacle location.

[0022] Furthermore, the multi-step adjacency matrix refers to recording the set of nodes that each node can reach in a straight line without being blocked by obstacles, and recording the length from the starting point of the exploration.

[0023] Furthermore, the forward and reverse node distance index data matrix is and Record the node distance index of each node, where N W The side length of the map scale.

[0024] Furthermore, the adjustment function of the adaptive ant quantity strategy is M t (K), defined as follows:

[0025]

[0026] In the formula, K max Indicates the maximum number of iterations; K indicates the number of current iterations.

[0027] Furthermore, the heuristic function ε composed of the node distance index and the Euclidean distance between the current node and the node to be transferred is ij , taking forward exploration as an example, the definition is:

[0028]

[0029] stES(allowed i )={ES(s)|s∈allowed i}.

[0030] Where β1, β2 and β3 are the probability transfer influencing factors, and their sizes are constants; ES(j) is The node distance index corresponding to the middle node j; K is the current iteration number; M is the number of ants in each iteration; minES (allowed i ) refers to the set ES(allowed i ), the minimum value allowed i is the feasible node set of node i; d ij Represents the distance from the current node i to the next node j, calculated as follows:

[0031]

[0032] In the formula, (x i ,y i ) is the grid rectangular coordinate of node i, (x j ,y j ) is the grid rectangular coordinate of node j.

[0033] Furthermore, the optimized probability transfer formula P ij , the definition is as follows:

[0034]

[0035] Where α is the influence factor of the heuristic function; allowed i It represents the set of all nodes that ant k can choose when selecting a node.

[0036] The convergence analysis of the cluster collaborative ant colony algorithm of the present invention is proved as follows:

[0037] It is sufficient to prove that the path length after each iteration of the algorithm tends to decrease or remain unchanged. W is the side length of the map scale; and is the forward and reverse node distance index data matrix, SE(i) and ES(i) are the forward and reverse node distance index of node i, respectively, which refers to the known shortest path distance from node i to the end point and the starting point. The proof is as follows:

[0038] (1) First, we analyze the algorithm’s probability selection problem and prove how the algorithm selects the optimal node when transferring nodes.

[0039] According to the heuristic function ε ij and the probability transfer function P ijIt can be obtained that as the number of iterations K increases, the probability of selecting the optimal node is as follows:

[0040]

[0041] Among them allowed i is the list of transferable nodes of node i, min refers to the set ES (allowed i ). That is, when selecting node transfer, the random probability selection in the early exploration is to make the algorithm's solution more diverse, ensure the generation of a global solution, and prevent the emergence of a local optimal solution. Nodes with known minimum path node distance indexes gradually show a higher probability of being selected. When the number of iterations K reaches a certain number, this probability will approach 1. At this time, when the algorithm constructs the node transfer of the path solution, each path node selected is to make the total path shortest, achieving convergence of the algorithm results.

[0042] (2) When the probability of selecting the optimal node approaches 1, analyze how to converge to the optimal solution.

[0043] When transferring to a node, the estimated shortest path is where path * is the shortest path in the forward search, is the path length from the current path planning to node i, ES(i) is the known shortest distance from node i to the end point. Continue to transfer nodes according to the optimal transfer method. Assuming that the next transfer node is j, there will be From this we can know the estimated shortest path path * It will only be less than or equal to the path length of the last reverse planning. When it is less than the path length of the last reverse planning, the path nodes are recorded and the forward node distance index matrix is ​​updated.

[0044] Therefore, when the next reverse search plan is made, it will proceed according to the guidance of the forward node distance index. At this time, the estimated shortest path length of the reverse plan will only be less than or equal to the path length of the last forward plan, that is, the following formula is used: path'≤path * , the equal sign exists when the current planning is the same as the previous forward planning path. Path' is the shortest path for reverse search. Similarly, as the number of iterations increases and the forward and reverse searches alternate, there will be a path in the forward planning. * ≤path', path'≤path in reverse planning * Until the forward and reverse planned paths are always consistent, the forward and reverse node distance indices are no longer updated, and path' = path * , the algorithm converges to the final result.

[0045] The time and space complexity analysis of the cluster collaborative ant colony algorithm of the present invention is as follows:

[0046] (1) Time complexity

[0047] The time complexity of an algorithm is a function of the input size n, which describes the number of basic operations the algorithm takes to complete the corresponding computing task when the input size is n. It can also intuitively reflect the performance and convergence speed of the algorithm. Based on the description of the process structure of the cluster collaborative ant colony algorithm of the present invention, its time complexity is analyzed step by step, as shown below:

[0048] 1) The time complexity of a single ant constructing a solution in one exploration is O(n 2 );

[0049] 2) The time complexity of initializing the bidirectional node distance index matrix is ​​O(Mn 2 );

[0050] 3) The time complexity of the interactive iteration of the collaborative ant colony algorithm is O(K max Mn 2 );

[0051] Where K max is the upper limit of the number of iterations preset by the algorithm, M is the number of ants in each iteration. When n is large enough, the influence of low-order powers can be ignored. From the above analysis, we can see that the time complexity of the overall algorithm can be expressed as T(n)=O(Mn 2 +K max Mn 2 )=O((K max +1)Mn 2 ).

[0052] (2) Space complexity

[0053] Space complexity is a measure of the amount of extra storage space that an algorithm temporarily occupies during its operation. In order to record the bidirectional node distance index data, two one-dimensional arrays of order n are required, and the space complexity is S(n) = O(2n); the call of the adjacency matrix requires a two-dimensional array of order n, and the space complexity is S(n) = O(n 2 ); In the process of the algorithm, a one-dimensional array of order n is required to record the path nodes, and the space complexity is S(n) = O(n); In order to record the path planning results in each iteration, a K max The matrix and cell array of size ×M respectively record the path length and the nodes of each path, and the space complexity is S(n) = O(K max M) and S(n) = O(K max Mn). From the above analysis, we can see that the space complexity of the overall algorithm can be expressed as S(n) = O(n 2+3n+K max M(n+1)).

[0054] The beneficial technical effects achieved by the present invention using the above technical scheme are as follows: the cluster collaborative ant colony algorithm of the present invention can adapt to the path planning problem in the field of unmanned driving, the adaptive step size optimization strategy can accurately avoid unnecessary node selection, and guide the algorithm to find a shorter path. The use of the two-way collaborative ant colony strategy effectively prevents the algorithm from falling into the local optimal solution, promotes the global progressive convergence of the algorithm, and significantly improves the exploration and convergence speed of the algorithm. Combined with the application of the adaptive ant colony number strategy and the reconstruction heuristic function, the stability of the algorithm is further enhanced, the operation time is significantly shortened, and the overall improvement of the algorithm performance is achieved, which has great practical value. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] The present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0056] Figure 1 This is a schematic diagram for comparing the multi-step improvements of the cluster collaborative ant colony algorithm;

[0057] Figure 2 It is a flowchart for the specific implementation of the cluster collaborative ant colony algorithm;

[0058] Figure 3 Selection of parameter values ​​for traditional ant colony algorithm;

[0059] Figure 4 Select the numerical values ​​of the parameters for the cluster collaborative ant colony algorithm;

[0060] Figure 5 This is the simulation result diagram of cluster collaborative ant colony algorithm experiment;

[0061] Figure 6 It is a comparison chart of iterative convergence between traditional ant colony algorithm and cluster cooperative ant colony algorithm;

[0062] Figure 7 This is a comparison chart of experimental simulation data results of traditional ant colony algorithm and cluster collaborative ant colony algorithm. DETAILED DESCRIPTION

[0063] In order to make the purpose, technical scheme and advantages of the present invention more obvious, the exemplary embodiments according to the present invention will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments of the present invention, and it should be understood that the present invention is not limited to the exemplary embodiments described herein. Based on the embodiments of the present invention described in the present invention, all other embodiments obtained by those skilled in the art without creative work should fall within the protection scope of the present invention.

[0064] It is further explained that the technical terms or scientific terms used in one or more examples of this specification should have the usual meanings understood by persons with ordinary skills in the field to which the present disclosure belongs. The words "first" and "first" used in the examples of this specification do not indicate any order, quantity or importance, but indicate the distinction between different components; "include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, but do not exclude other elements or objects; "up", "down", "left", "right" and the like are only used to indicate relative position relationships. When the absolute position of the described object changes, the relative position relationship may also change accordingly; the relevant data mentioned in this specification and their expression methods only represent a recording method and are not limited to the described example embodiments.

[0065] As attached Figure 2 As shown, the cluster collaborative ant colony algorithm for solving the unmanned driving path planning problem provided in this embodiment includes the following specific steps:

[0066] Step 1: Receive the path planning request of the driverless vehicle and obtain map-related information through the Internet of Vehicles function.

[0067] Step 2: Create a grid map and obtain a multi-step adjacency matrix based on the adaptive step strategy.

[0068] Step 3: Initialize parameters, such as the heuristic function influence factor α, the probability transfer influence factor β, the number of ants M, the number of iterations K, and the node distance index table, and divide the ant colony into the forward population and the reverse population.

[0069] Step 4: Use the traditional ant colony algorithm to perform preliminary path planning, the purpose is to update the node distance index data matrix of the forward and reverse populations, and provide a data basis for the cluster collaborative ant colony algorithm.

[0070] Step 5: According to the adaptive ant number function m t (K) Adjust the number of ants currently exploring and divide the ant colony into two populations, forward and reverse, according to the heuristic function ε dominated by the node distance index ij and the probability transfer function P ij Perform node transfer and introduce probabilistic expansion of optimization processing.

[0071] Step 6: After the node is transferred, record the node it has moved to and update the node distance index data matrix.

[0072] Step 7: After completing the exploration, determine whether m is the maximum number of ants in the current iteration. If so, jump to step 8; otherwise, set m=m+1 and execute step 5.

[0073] Step 8: Determine whether the number of iterations k reaches the maximum number of iterations K max If yes, jump to step 9; otherwise, set k=k+1 and execute step 5.

[0074] Step 9: Complete the cluster collaborative ant colony algorithm path planning and output the optimal path.

[0075] Furthermore, in step 4, the traditional ant colony algorithm is used to perform path planning in the preliminary stage. The parameters of the traditional ant colony algorithm are set as shown in the attached figure. Figure 3 shown.

[0076] Furthermore, the multi-step adjacency matrix refers to recording the set of nodes that each node can reach in a straight line without being blocked by obstacles and recording the length of the distance.

[0077] Furthermore, the forward and reverse node distance index data matrix is and Record the node distance index of each node, where N W The side length of the map scale.

[0078] Furthermore, the adjustment function of the adaptive ant quantity strategy is M t (K), defined as follows:

[0079]

[0080] In the formula, K max Indicates the maximum number of iterations; K indicates the number of current iterations.

[0081] Furthermore, the heuristic function ε composed of the node distance index and the Euclidean distance between the current node and the node to be transferred is ij , taking forward exploration as an example, the definition is:

[0082]

[0083] stES(allowed i )={ES(s)|s∈allowed i}.

[0084] Where β1, β2 and β3 are the probability transfer influencing factors, and their sizes are constants; ES(j) is The node distance index corresponding to the middle node j; K is the current iteration number; M is the number of ants in each iteration; minES (allowed i ) refers to the set ES(allowed i ), the minimum value allowed i is the feasible node set of node i; d ijRepresents the distance from the current node i to the next node j, calculated as follows:

[0085]

[0086] In the formula, (x i ,y i ) is the grid rectangular coordinate of node i, (x j ,y j ) is the grid rectangular coordinate of node j.

[0087] Furthermore, the optimized probability transfer formula P ij , the definition is as follows:

[0088]

[0089] Where α is the influence factor of the heuristic function; allowed i It represents the set of all nodes that ant k can choose when selecting a node.

[0090] Furthermore, the cluster collaborative ant colony algorithm provided in this embodiment for solving the unmanned driving path planning problem has the following parameter settings: Figure 4 The simulation results of 20×20 scale map and 30×30 scale map are shown in the attached figure. Figure 5 As shown in (a) and (b), the iterative convergence of the traditional ant colony algorithm and the cluster cooperative ant colony algorithm of the present invention in the 20×20 scale map and the 30×30 scale map is shown in the attached figure. Figure 6 (a) and (b). The simulation results show that the performance of this algorithm is better than the traditional ant colony algorithm, and it shows higher stability and efficiency in terms of optimization effect and computing time.

[0091] In this example, the simulation results of unmanned driving path planning based on the traditional ant colony algorithm and the cluster collaborative ant colony algorithm of the present invention are compared as shown in the attached figure. Figure 7 shown.

Claims

1. A method for unmanned driving path planning based on cluster collaborative ant colony algorithm, characterized in that: Here are the steps: Step 1: Introduce an adaptive step-size strategy to expand the node selection set of the ant colony during node selection and reduce the selection of redundant nodes; Step 2: Introduce an adaptive ant colony quantity strategy to dynamically adjust the number of ant colonies during the iteration process; Step 3: Introduce a two-way ant colony collaboration strategy and record the node distance index of each node. Ants can make node transfer decisions by evaluating the node distance index of each node. The node distance index and the Euclidean distance between the current node and the node to be transferred together constitute the heuristic function of the algorithm. Step 4: Replace the pheromone function and rebuild the heuristic function ε dominated by the node distance index ij .

2. The adaptive step size strategy according to claim 1, characterized in that: The feasible node selection without limiting the step range makes the ant's single movement distance and movement direction more flexible. The node selection logic of the adaptive step ant colony algorithm is to use the nodes that can be reached in a straight line from the current node and are not blocked by obstacles as the optional nodes for the next movement. The increase in the number of single search neighborhoods can reduce the total number of searches of the algorithm and enhance the algorithm's optimization ability.

3. The adaptive ant colony quantity strategy described in claim 1 is characterized in that: In order to improve the stability and global optimization ability of the algorithm and avoid falling into the local optimal solution, the present invention introduces an adaptive ant colony quantity strategy to dynamically adjust the number of ant colonies in the iterative process. The adjustment function expression is: In the formula, K max is the upper limit of the maximum number of iterations; K is the current number of iterations. According to the dynamic population size adjustment, the ant colony algorithm can fully traverse the map in the early exploration and obtain more possible solutions. In the later process, the number of ant colonies can be reduced to avoid the concentration of late convergence results and waste of some computing resources.

4. The bidirectional ant colony collaboration strategy described in claim 1 is characterized in that: The number of ants in the current iteration is divided into two sub-populations, namely the forward population and the reverse population. The two populations are made to explore alternately from the starting point to the end point and from the end point to the starting point. In the process of moving forward, each ant will record the distance from each node it has passed to the starting point. This data is named the node distance index and recorded in the node distance index data set of the corresponding sub-population. After the ant node is transferred in each iteration, the current node distance index is compared with the size of the corresponding data in the data set. If it is smaller, the data is replaced, otherwise it is not changed. When selecting nodes, the forward sub-population uses the node distance index recorded by the reverse sub-population as a guide to select the optimal path. Similarly, the reverse sub-population uses the node distance index recorded by the forward sub-population as a guide. The two populations feedback and cooperate with each other and finally converge to obtain the optimal path.

5. The heuristic function based on the node distance index as reconstructed in claim 1 is characterized in that: In order to improve the efficiency of the optimization algorithm and accelerate the convergence speed, the present invention reconstructs the heuristic function ε dominated by the node distance index ij , taking forward exploration as an example, its expression is: s.t.ES(allowed i )={ES(s)|s∈allowed i }. In the formula, ε ij is the heuristic function from node i to node j; β1, β2 and β3 are the probability transfer influencing factors; ES(j) is The node distance index corresponding to the middle node j; K is the current iteration number; M is the number of ants in each iteration; minES(allowed i ) refers to the set ES(allowed i ), the minimum value allowed i is the feasible node set of node i. The node distance index of the bidirectional ant colony is used to replace pheromone to construct a new heuristic function. In order to deal with the uncertainty in the transfer node, the probability extension of the optimization process is introduced. By optimizing the node with the smallest cost, the algorithm gradually shows a higher probability of being selected as the number of iterations increases, making the ant colony exploration more guided, thereby accelerating the convergence of the algorithm.

6. The node distance index according to claim 4, characterized in that: The node distance index is a characteristic value assigned to each node to provide guidance for ants. Ants can make node transfer decisions by evaluating the characteristic value of each node. The forward and reverse node distance index data matrix is and Where N W The side length of the map scale.