Unmanned aerial vehicle path planning method based on multi-strategy improved ant colony algorithm

By introducing angle functions and adaptive pheromone volatile factors into the ant colony algorithm, the pheromone update rules are optimized, and parameters are adjusted through pre-experimental adjustment, the slow convergence speed and path turning points of the traditional ant colony algorithm in drone path planning are solved. More and local optimal solutions have been achieved, and more efficient and safe path planning has been achieved.

CN119024859AActive Publication Date: 2024-11-26HEILONGJIANG HONGYUE SURVEYING & MAPPING TECHNOLOGY SERVICE CO LTD

Patent Information

Application Number
CN202411054465.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-02
Publication Date
2024-11-26
Estimated Expiration
2044-08-02

AI Technical Summary

Technical Problem

In the path planning of drone, traditional ant colony algorithm has problems such as slow convergence speed, many path turning points, and easy to fall into local optimal solutions and complex parameter adjustment, which affects its efficiency and safety in complex environments.

Method used

By introducing angle functions and adaptive pheromone volatile factors, the pheromone update rules are optimized, and parameters are adjusted through pre-experimental adjustments, the state transition probability and pheromone update strategy of the ant colony algorithm are improved to reduce random searches, reduce inflection points, and enhance diversity and stability to avoid local optimal solutions.

Benefits of technology

The convergence speed and path optimization performance of the ant colony algorithm are improved, allowing the drone to reach the target position faster and accurately, reduce energy loss, and show higher robustness and safety in complex environments.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure BDA0004975933160000022
    Figure BDA0004975933160000022
  • Figure BDA0004975933160000031
    Figure BDA0004975933160000031
  • Figure BDA0004975933160000032
    Figure BDA0004975933160000032
Patent Text Reader

Abstract

The invention discloses an unmanned aerial vehicle path planning method based on a multi-strategy improved ant colony algorithm, and relates to the technical field of unmanned aerial vehicle path planning. In order to solve the problems of slow convergence speed, easy falling into a local optimal solution, complex parameter adjustment and the like of a traditional ant colony algorithm in unmanned aerial vehicle path planning, the invention provides an unmanned aerial vehicle path planning method based on a multi-strategy improved ant colony algorithm, and the method comprises the steps: introducing an angle function into a state transition probability to provide directional information; therefore, the purposes of accelerating the convergence speed and reducing corners are achieved; a strategy of expanding pheromones inwards and limiting the concentration of the pheromones is adopted in a pheromone updating rule, so that the problems of premature convergence and falling into a local optimal solution of the algorithm are avoided; according to the method, pheromone volatilization factors are dynamically adjusted in a self-adaptive mode, values of corresponding parameters are dynamically adjusted through pre-experiment processing and target function design, an optimal parameter combination is obtained, and therefore the convergence effect of the algorithm is improved. According to the method, various performances in the unmanned aerial vehicle path planning process can be improved, and the method is more suitable for a large-scale complex environment.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicle path planning, and in particular to a method for unmanned aerial vehicle path planning based on a multi-strategy improved ant colony algorithm. Technical Background

[0002] Due to its strong autonomous capabilities and high flexibility, drones have become an important tool to meet human needs and facilitate human work in today's production and life. The current research and development of drones has not only promoted the progress of many fields such as industrial automation and agricultural intelligent production, but also provided convenient services in daily medical rescue and cargo transportation. Perception and autonomous navigation technology are the core research areas of drones, which enable drones to autonomously plan action routes that meet needs based on environmental information obtained by sensors. Therefore, path planning, as a key link in autonomous navigation, has become a research hotspot in the field of drones.

[0003] Path planning refers to the process of finding a path from the starting point to the target point safely while avoiding obstacles, and optimizing its performance as much as possible. Typical path planning algorithms include Dijkstra algorithm, A* algorithm, artificial potential field method, rapidly expanding random tree algorithm, genetic algorithm, ant colony algorithm, particle swarm algorithm, etc. However, in the process of path planning, drones need to consider static known obstacles, map environment complexity, target point location and other information to plan the global path. As a global path planning algorithm using distributed computing, the ant colony algorithm has the advantage of strong robustness. Compared with other algorithms, it has low requirements for the initial route; it is a self-organizing algorithm with strong adaptability to different problem environments; it has parallelism, strong global search capabilities, and is easy to integrate with other algorithms.

[0004] However, the traditional ant colony algorithm still has the following shortcomings in the global path planning process:

[0005] (1) Slow convergence: In the initial stage of the algorithm, the probability of ants choosing all paths is the same, so the search process is highly random and blind. Especially when dealing with complex or large-scale problems, the search efficiency is too low.

[0006] (2) Too many turning points: The traditional ant colony algorithm does not consider the size of the corners and the number of turning points, which will lead to excessive energy loss and low safety factor in actual application.

[0007] (3) Easy to fall into local optimal solutions: Due to the positive feedback mechanism of the algorithm, the subsequent search path is concentrated on some candidate solutions, which reduces the diversity of the population and is not conducive to finding the global optimal solution.

[0008] (4) Complex parameter adjustment: The setting of parameters has a great impact on the performance of the algorithm, but there is no perfect theory to determine the value of the ant colony algorithm parameters. Usually, we can only rely on a large number of experiments and experience to give the optimal parameter combination. Summary of the invention

[0009] In order to solve the above-mentioned problems of the ant colony algorithm from multiple angles, the main purpose of the present invention is to propose a UAV path planning method based on a multi-strategy improved ant colony algorithm, which improves the convergence speed of the ant colony algorithm, reduces the path length, and optimizes its various performances by improving the state transition probability, pheromone update rules and parameter adjustment.

[0010] In order to achieve the above-mentioned purpose, the present invention provides a UAV path planning method based on a multi-strategy improved ant colony algorithm. Aiming at the problem that the search process in the traditional ant colony algorithm is blind and there are many turning points in the planning path, the present invention introduces an angle function containing two factors, namely the steering angle and the corner angle, in the state transition probability to provide the ants with the direction to the target point, reduce unnecessary random searches, speed up the convergence speed, and simultaneously achieve the purpose of reducing turning points and corners.

[0011] Furthermore, in order to improve the algorithm search effect, the present invention adjusts the pheromone volatilization factor into an adaptive parameter so that it changes with the number of iterations and obeys a normal distribution, thereby ensuring the stability of the algorithm while increasing the diversity and flexibility of the search.

[0012] Furthermore, based on the pheromone update rules of the traditional algorithm, the present invention introduces the strategy of expanding pheromones inward and limiting pheromone concentration. While narrowing the search range of the optimal path, it is beneficial for ants to find a better path in subsequent searches and avoids directly converging on a non-optimal path due to the accumulation of pheromones to form a local optimal solution.

[0013] Finally, in order to address the problem that the ant colony algorithm has a strong dependence on parameters but a complex parameter adjustment process, the present invention conducts a preliminary experiment to dynamically adjust the values ​​of relevant parameters according to the objective function, and uses the parameter values ​​obtained at the end of the experimental run as the optimal parameter combination of the present invention.

[0014] The technical solution adopted by the present invention is:

[0015] A UAV path planning method based on a multi-strategy improved ant colony algorithm mainly includes the following steps:

[0016] Step 1: Create a two-dimensional grid map and a three-dimensional space map to determine the positions of the starting point S, the target point G, and all obstacles;

[0017] Step 2: Use the traditional ant colony algorithm to perform preliminary path planning, and provide an algorithm platform for the UAV path planning method based on the multi-strategy improved ant colony algorithm;

[0018] Step 3: Introduce the angle function into the state transition probability and its weight γ, reducing unnecessary random searches and effectively solving the problem of blind search process and many turning points in the planning path;

[0019] Step 4: dynamically and adaptively adjust the pheromone volatilization factor ρ, so that the algorithm can continuously adjust the influence of pheromone on path selection as the number of iterations changes, improve the search effect at different stages of the algorithm, and increase the diversity of search while maintaining stability;

[0020] Step 5: Introduce the strategy of inward expansion of pheromone in the pheromone update rule, and perform secondary update on the pheromone concentrations between feasible points in some preferred areas to avoid premature convergence of the algorithm and fall into local optimum, and ensure the quality of subsequent search paths;

[0021] Step 6: Limit the updated residual pheromone concentration to [τ min ,τ max ] range, to avoid excessive pheromone concentration on a certain path, which would cause the path to be suboptimal but the algorithm would converge because the pheromone concentration is much greater than other paths;

[0022] Step 7, based on the above steps, a preliminary experiment is performed on the parameters α, β and γ, and dynamic adjustments are made according to the objective function. The values ​​of the three parameters at the end of the operation are used as the values ​​of the weight coefficients of the present invention, avoiding the complex process of determining the parameters through multiple experiments.

[0023] Furthermore, in order to speed up the convergence of the algorithm, the present invention introduces an angle function in step 3: The definition is as follows:

[0024]

[0025] The steering angle θ jg and the corner angle θ sj The steering angle is formed by connecting the current node with the next node and the target point respectively. The smaller the angle, the closer the direction of movement to the next node is to the target point, which provides a certain directionality for the ants, thereby reducing unnecessary random searches and accelerating convergence. The corner angle is formed by connecting the current node with the previous node and the next node respectively. The larger the angle, the smoother the path is, which can achieve the purpose of reducing turning points and corners.

[0026] Furthermore, after the present invention introduces the angle function and its weight, the state transition probability calculation formula is as follows:

[0027]

[0028] Among them, τ ij (t) is the pheromone concentration between node i and node j at time t, α is the pheromone weight, η ij (t) is the heuristic information representing the expected degree, β is the heuristic information weight, γ is the angle function weight, J k (i) is the set of nodes that ant k can visit in the next step.

[0029] Furthermore, in order to improve the search effect at different stages of the algorithm, the present invention introduces an adaptive pheromone volatilization factor in step 4, which is defined as follows:

[0030]

[0031] Among them, σ is the scale parameter, a is the adjustment coefficient, N c is the current iteration number, N max is the maximum number of iterations. In the early stage of the algorithm, the ρ value is small, which can prevent excessive pheromone volatilization and weaken the effect of pheromones; in the middle stage of the algorithm, the ρ value is large, which will increase the diversity of the search and prevent the algorithm from converging prematurely due to the accumulation of pheromones; in the later stage, the ρ value is small, which will increase the guidance of pheromones and make the algorithm converge to the shortest path.

[0032] Furthermore, after the present invention introduces the adaptive pheromone volatilization factor, the pheromone update rule is as follows:

[0033]

[0034] Where m is the number of ants, is the pheromone concentration released by ant k between nodes i and j in this iteration, and its expression is

[0035]

[0036] Among them, Q is the total amount of pheromones released by ants during one traversal, L k is the length of the path traversed by ant k in this iteration.

[0037] Furthermore, in order to prevent the algorithm from falling into a local optimal solution, the present invention introduces a strategy of expanding pheromones inward in step 5. First, the ideal path is defined as the shortest path from the starting point to the target point when there are no obstacles. The preferred path is taken as the path shorter than the average path length in this iteration. The area where the preferred path is expanded inward to the ideal path is called the preferred area. The present invention performs a secondary update on the basis of the traditional algorithm pheromone update rule, and increases the pheromone concentration between feasible points in some preferred areas. The calculation formula is as follows:

[0038]

[0039] Where L avg is the average length of all paths in this iteration, L i The path length is less than the average value in this iteration. This narrows the search range of the optimal path and helps ants find a better path in the area in subsequent searches, avoiding the formation of a local optimal solution on a non-optimal preferred path due to the accumulation of pheromones.

[0040] Furthermore, considering that there may be too many obstacles in the preferred area, it will be impossible to form a feasible path, causing the ants to stagnate or bypass obstacles to form a longer path. The present invention determines whether to update the internal pheromone by judging the proportion of obstacles in the preferred area. If the proportion is small, it indicates that there may be a shorter path inside, and the pheromone in the area is updated; otherwise, it is not updated. That is, the secondary update of pheromones must meet the following constraints at the same time:

[0041]

[0042] Where n is the limit of the number of iterations, N obs N is the number of grids occupied by obstacles in the formed preferred area. all is the total number of grids in the preferred area, and δ is the threshold of obstacle ratio.

[0043] Furthermore, in order to avoid excessive accumulation of pheromones leading to non-optimal path convergence, the present invention limits the updated residual pheromone concentration to the following range in step 6:

[0044]

[0045] where τ min is the minimum value of pheromone concentration, τ max is the maximum pheromone concentration.

[0046] Furthermore, since there is no perfect theory to determine the values ​​of the ant colony algorithm parameters, the optimal parameter combination is usually given based on multiple experimental comparisons. Therefore, the present invention performs a preliminary experimental treatment on the weight parameters α, β and γ in step 7, and dynamically adjusts the values ​​of the relevant parameters by designing the objective function. At the end of the operation, the obtained parameter values ​​are used as the optimal parameter combination of the improved ant colony algorithm. First, the weight parameters are set to α=1, β=1, γ=1, and then the corresponding parameters are adjusted according to the following conditions:

[0047] The pheromone weight α determines the degree of influence of the residual pheromone concentration on the path selection. Therefore, the present invention dynamically adjusts the parameter α according to the variance change of the path length after each iteration. When the variance of this iteration becomes larger than that of the previous one, it indicates that the search process is more random, and the value of α is increased at this time; when the variance becomes smaller, it indicates that there are fewer search paths and it is easy to produce a local optimal solution, and the value of α is reduced at this time. That is, the objective function is

[0048] f(k)=S(k)-S(k-1), k≥2

[0049] Where S(k) is the variance of all path lengths at the kth iteration. The dynamic change rule of the corresponding parameter α is as follows:

[0050]

[0051] Similarly, the heuristic information weight β determines the degree of influence of the distance of the next move on the path selection. Therefore, the present invention dynamically adjusts β according to the shortest path length after each iteration, and the corresponding objective function is:

[0052] g(k)=L min (k)-L min (k-1),k≥2

[0053] Where L min (k) is the shortest path length in the kth iteration. The dynamic change rule of the corresponding parameter β is as follows:

[0054]

[0055] The angle function weight γ determines the influence of the angle information of the next node on the path selection. Therefore, the present invention dynamically adjusts γ according to the angle deviation between the shortest path and the ideal path after each iteration. The corresponding objective function is:

[0056] h(k)=θ(k)-θ(k-1), k≥2

[0057] Where θ(k) is the sum of the angles formed by the target point and the lines connecting the turning points and the starting point of the shortest path in the kth iteration. The dynamic change rules of the corresponding parameter γ are as follows:

[0058]

[0059] Compared with the prior art, the present invention has the following beneficial effects:

[0060] The present invention adopts the above technical solution, so that when the UAV performs path planning in a global static environment, the algorithm is provided with direction information pointing to the target point, which reduces unnecessary random searches, speeds up the convergence speed compared to traditional algorithms, and has more advantages in large-scale complex environments. Secondly, the secondary update and the introduction of the adaptive pheromone volatilization factor in the pheromone update rule solve the problem that the traditional ant colony algorithm is prone to fall into the local optimum, and optimizes the search effect of the algorithm at different stages. Finally, the problem of complex parameter adjustment is solved through pre-experimental processing. Therefore, the present invention improves the performance of the ant colony algorithm from many aspects, enables the UAV to reach the target position quickly and accurately, and reduces its energy consumption. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] The present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0062] Figure 1 This is a flowchart of a UAV path planning method based on a multi-strategy improved ant colony algorithm;

[0063] Figure 2 This is a diagram of preliminary experimental results based on the improved ant colony algorithm of the present invention;

[0064] FIG3(a) and FIG3(b) are simulation diagrams of two-dimensional path planning results based on the traditional ant colony algorithm and the improved ant colony algorithm;

[0065] Figure 4 This is a comparison diagram of the convergence of two-dimensional path planning based on traditional ant colony algorithm and improved ant colony algorithm;

[0066] Figure 5 The simulation diagram of the three-dimensional path planning results based on the traditional ant colony algorithm and the improved ant colony algorithm;

[0067] Figure 6 This is a comparison diagram of the convergence of three-dimensional path planning based on traditional ant colony algorithm and improved ant colony algorithm. DETAILED DESCRIPTION

[0068] In order to make the improved purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the following embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making creative work are within the scope of protection of the present invention.

[0069] At present, when performing path planning in a global static environment, the traditional ant colony algorithm has the following shortcomings: the early search process of the algorithm is highly random, resulting in slow convergence of the algorithm, long search time, and many turning points; the algorithm is prone to concentrate the search path on some candidate solutions under the action of the positive feedback mechanism, thereby reducing the diversity of the population and causing the algorithm to fall into a local optimal solution; the weight parameters in the algorithm have a great influence on the experimental results, but usually can only rely on a large number of experiments and experience to determine the value, and the parameter adjustment process is relatively complex. Based on this, the main purpose of this application is to propose a UAV path planning method based on a multi-strategy improved ant colony algorithm, which solves the problems of slow convergence, many turning points, easy to fall into a local optimal solution, and complex parameter adjustment when using the traditional ant colony algorithm for path planning by improving the state transition probability, pheromone update rules, and parameter adjustment.

[0070] To facilitate understanding of this embodiment, a drone path planning method based on a multi-strategy improved ant colony algorithm disclosed in an embodiment of the present invention is first introduced in detail.

[0071] Example:

[0072] A UAV path planning method based on a multi-strategy improved ant colony algorithm mainly includes the following steps:

[0073] Step 1: Create a two-dimensional grid map and a three-dimensional space map to determine the positions of the starting point S, the target point G, and all obstacles;

[0074] Step 2: Use the traditional ant colony algorithm to perform preliminary path planning, and provide an algorithm platform for the UAV path planning method based on the multi-strategy improved ant colony algorithm;

[0075] Step 3: Introduce the angle function into the state transition probability and its weight γ, reducing unnecessary random searches and effectively solving the problem of blind search process and many turning points in the planning path;

[0076] Step 4: dynamically and adaptively adjust the pheromone volatilization factor ρ, so that the algorithm can continuously adjust the influence of pheromone on path selection as the number of iterations changes, improve the search effect at different stages of the algorithm, and increase the diversity of search while maintaining stability;

[0077] Step 5: Introduce the strategy of inward expansion of pheromone in the pheromone update rule, and perform secondary update on the pheromone concentrations between feasible points in some preferred areas to avoid premature convergence of the algorithm and fall into local optimum, and ensure the quality of subsequent search paths;

[0078] Step 6: Limit the updated residual pheromone concentration to [τ min ,τ max] range, to avoid excessive pheromone concentration on a certain path, which would cause the path to be suboptimal but the algorithm would converge because the pheromone concentration is much greater than other paths;

[0079] Step 7, based on the above steps, a preliminary experiment is performed on the parameters α, β and γ, and dynamic adjustments are made according to the objective function. The values ​​of the three parameters at the end of the operation are used as the values ​​of the weight coefficients of the present invention, avoiding the complex process of determining the parameters through multiple experiments.

[0080] Furthermore, in step 1, MATLAB R2016b simulation software is used to establish a two-dimensional grid map and a three-dimensional space map, and the positions of the starting point S, the target point G and all obstacles are determined.

[0081] Furthermore, the two-dimensional path planning result obtained by using the traditional ant colony algorithm in step 2 is shown in Figure 3(a).

[0082] Furthermore, the angle function is introduced in step 3 The definition is as follows:

[0083]

[0084] where θ jg is the steering angle, which is formed by connecting the current node with the next node and the target point respectively; θ sj is the corner angle, which is formed by connecting the current node with the previous node and the next node respectively.

[0085] Furthermore, based on the above improvements, the state transition probability calculation formula is as follows:

[0086]

[0087] Among them, τ ij (t) is the pheromone concentration between node i and node j at time t, α is the pheromone weight, η ij (t) is the heuristic information representing the expected degree, β is the heuristic information weight, γ is the angle function weight, J k (i) is the set of nodes that ant k can visit in the next step.

[0088] Furthermore, in step 4, an adaptive pheromone volatility factor is introduced, which is defined as follows:

[0089]

[0090] Among them, σ is the scale parameter, a is the adjustment coefficient, N c is the current iteration number, N max is the maximum number of iterations, in this embodiment, N max =100.

[0091] Furthermore, based on the above improvements, the pheromone update rules are as follows:

[0092]

[0093] Where m is the number of ants, is the pheromone concentration released by ant k between nodes i and j in this iteration, and its expression is

[0094]

[0095] Among them, Q is the total amount of pheromones released by ants during one traversal, L k is the length of the path traversed by ant k in this iteration. In this embodiment, m=50 and Q=1.

[0096] Furthermore, the strategy of expanding pheromones inward is introduced in step 5. First, it is determined whether the preferred area meets the following constraints:

[0097]

[0098] Where n is the limit of the number of iterations, N obs N is the number of grids occupied by obstacles in the formed preferred area. all is the total number of grids in the preferred area, and δ is the threshold value of the obstacle ratio. In this embodiment, n=40 and δ=0.25.

[0099] If satisfied, the pheromones between the feasible points in the preferred area are updated twice as follows:

[0100]

[0101] Where L avg is the average length of all paths in this iteration, L i are the paths whose path length is less than the average value in this iteration.

[0102] Furthermore, in step 6, the updated residual pheromone concentration is limited to the following range:

[0103]

[0104] where τ min is the minimum value of pheromone concentration, τ max is the maximum pheromone concentration.

[0105] Furthermore, in step 7, a preliminary experiment was conducted on the weight parameters α, β and γ. First, the weight parameters were set to α = 1, β = 1, γ = 1, and the objective functions corresponding to the design parameters α, β, and γ were

[0106] f(k)=S(k)-S(k-1), g(k)=L min (k)-L min (k-1),h(k)=θ(k)-θ(k-1)

[0107] Where k ≥ 2, S(k) is the variance of all path lengths at the kth iteration, L min (k) is the length of the shortest path in the kth iteration, and θ(k) is the sum of the angles formed by the target point and the lines connecting the turning points and the starting point of the shortest path in the kth iteration.

[0108] The relevant parameters are dynamically adjusted according to the changes in the objective function after each iteration. The dynamic change rules are as follows:

[0109]

[0110] At the end of the run, the obtained parameter values ​​are used as the optimal parameter combination of the improved ant colony algorithm.

[0111] The results obtained based on the above preliminary experiments are shown in the attached Figure 2 shown.

[0112] Furthermore, the two-dimensional path planning result obtained based on the improved ant colony algorithm is shown in FIG3(b), which is compared with the path planning convergence curve of the traditional ant colony algorithm. Figure 4 The three-dimensional path planning results based on the traditional ant colony algorithm and the improved ant colony algorithm are shown in the attached Figure 5 As shown in the figure, the path planning convergence curve is compared with the attached Figure 6 shown.

[0113] The following table compares the path planning simulation results in two-dimensional and three-dimensional environments based on the traditional ant colony algorithm and the improved ant colony algorithm proposed in the present invention.

[0114] Table 1 Comparison of simulation results of two-dimensional path planning before and after the ant colony algorithm improvement

[0115] algorithm Path length / m Running time / s Number of turns Iteration convergence times Traditional ant colony algorithm 46.6 26.54 15 38 Improved ant colony algorithm 43.3 33.93 7 8

[0116] Table 2 Comparison of 3D path planning simulation results before and after the ant colony algorithm improvement

[0117] algorithm Path length / m Iteration convergence times Number of turns Corner angle and Traditional ant colony algorithm 145.2 56 20 1770.4 Improved ant colony algorithm 131.3 25 19 1391.6

[0118] The simulation results of this embodiment show that when performing path planning in a two-dimensional environment, the improved ant colony algorithm proposed in the present invention has a faster convergence speed, fewer turns, and a significantly shorter optimal path length than the traditional algorithm; in a three-dimensional environment, the path length planned by the improved ant colony algorithm is optimized by 9.6% compared with the traditional algorithm, the number of iterative convergences is optimized by 55.4%, and the turning angle is reduced by 21.4% compared with the traditional algorithm, making the path smoother. Even though the running time of the improved algorithm is longer due to the secondary update of pheromones, the above simulation results show that the improved ant colony algorithm proposed in the present invention can converge to the global optimal solution faster, avoid the problem of local optimality, and plan a better path with better safety performance.

Claims

1. A UAV path planning method based on a multi-strategy improved ant colony algorithm, characterized in that: The following steps are involved: Step 1: Create a two-dimensional grid map and a three-dimensional space map to determine the positions of the starting point S, the target point G, and all obstacles; Step 2: Use the traditional ant colony algorithm to perform preliminary path planning, and provide an algorithm platform for the UAV path planning method based on the multi-strategy improved ant colony algorithm; Step 3: Introduce the angle function into the state transition probability and its weight γ, reducing unnecessary random searches and effectively solving the problem of blind search process and many turning points in the planning path; Step 4: dynamically and adaptively adjust the pheromone volatilization factor ρ, so that the algorithm can continuously adjust the influence of pheromone on path selection as the number of iterations changes, improve the search effect at different stages of the algorithm, and increase the diversity of search while maintaining stability; Step 5: Introduce the strategy of inward expansion of pheromone in the pheromone update rule, and perform secondary update on the pheromone concentrations between feasible points in some preferred areas to avoid premature convergence of the algorithm and fall into local optimum, and ensure the quality of subsequent search paths; Step 6: Limit the updated residual pheromone concentration to [τ min ,τ max ] range, to avoid excessive pheromone concentration on a certain path, which would cause the path to be suboptimal but the algorithm would converge because the pheromone concentration is much greater than other paths; Step 7, based on the above steps, a preliminary experiment is performed on the parameters α, β and γ, and dynamic adjustments are made according to the objective function. The values ​​of the three parameters at the end of the operation are used as the values ​​of the weight coefficients of the present invention, avoiding the complex process of determining the parameters through multiple experiments.

2. The method for UAV path planning based on multi-strategy improved ant colony algorithm as claimed in claim 1, characterized in that: In step 3, in order to speed up the convergence of the algorithm and reduce the corner, the present invention introduces a steering angle θ into the state transition probability. jg and the corner angle θ sj The angle function of two factors is defined as follows:

3. The method for unmanned aerial vehicle path planning based on multi-strategy improved ant colony algorithm as claimed in claim 1, characterized in that: In step 4, in order to improve the algorithm search effect, the present invention adjusts the pheromone volatility factor to a form that changes with the number of iterations and obeys a normal distribution, and its expression is as follows: Where σ is the scale parameter, a is the adjustment coefficient, N c is the current iteration number, N max is the maximum number of iterations.

4. The method for UAV path planning based on multi-strategy improved ant colony algorithm according to claim 1, characterized in that: In step 5, in order to prevent the algorithm from falling into a local optimum and to ensure the quality of the subsequent search path, the present invention introduces a strategy of expanding pheromones inward and limiting pheromone concentration. When the obstacle ratio is small in the preferred area formed by the path whose path length is less than the average value and the ideal path at the initial iteration, the pheromones between the feasible points in the corresponding preferred area are updated twice based on the traditional algorithm update rules as follows: In step 6, the residual pheromone concentration is limited to the following range: Where Q is the total amount of pheromones released by ants during one traversal, L avg is the average length of all paths in this iteration, L i is the path whose path length is less than the average value in this iteration, τ min is the minimum value of pheromone concentration, τ max is the maximum pheromone concentration.

5. The method for unmanned aerial vehicle path planning based on multi-strategy improved ant colony algorithm as claimed in claim 1, characterized in that: In step 7, in order to avoid complicated manual parameter adjustment, the present invention conducts a preliminary experiment, designs an objective function, dynamically adjusts the values ​​of relevant parameters, and uses the obtained parameter values ​​as the optimal parameter combination of the improved ant colony algorithm at the end of the operation. The objective functions corresponding to the parameters α, β, and γ are f(k)=S(k)-S(k-1), k≥2 g(k)=L min (k)-L min (k-1),k≥2 h(k)=θ(k)-θ(k-1), k≥2 Where S(k) is the variance of all path lengths at the kth iteration, L min (k) is the length of the shortest path in the kth iteration, and θ(k) is the sum of the angles formed by the target point and the lines connecting the turning points and the starting point of the shortest path in the kth iteration.

Citation Information

Patent Citations

  • Method of path planning based on improved ant colony algorithm

    CN107272679A

  • Unmanned aerial vehicle route planning method based on self-adaptive multi-modal fusion ant colony algorithm

    CN110319829A

  • Unmanned aerial vehicle route planning method based on improved ant colony algorithm

    CN114020045A

  • Unmanned aerial vehicle three-dimensional path planning method based on improved ant colony algorithm

    CN116578116A

  • Unmanned aerial vehicle three-dimensional path planning method based on improved ant colony algorithm

    CN117850448A

Cited By

  • Unmanned aerial vehicle path planning method and system based on adaptive Hybrid A* algorithm

    CN120063270A

  • Unmanned aerial vehicle cluster adaptive way-finding method and system based on bionic intelligence

    CN120428736A

  • Unmanned aerial vehicle path planning parameter optimization method and device, equipment and storage medium

    CN122366805A