Improved snake vulture optimization algorithm path planning method and system

Abstract

The invention discloses a path planning method and system for an improved snake vulture optimization algorithm, and relates to the technical field of path planning algorithms, and the method comprises the steps: generating a high-uniformity initial population through Tent and Bernoulli double-chaotic mapping, and enhancing the global exploration capability; secondly, a global exploration strategy of an eagle optimization algorithm is introduced, and path search robustness in a complex obstacle environment is enhanced; meanwhile, a self-adaptive T distribution disturbance strategy is adopted, the algorithm development and exploration capability is dynamically balanced, and local optimum is effectively avoided. After a workshop environment model is constructed based on a grid method, continuous domain codes are converted into discrete grid paths, and an optimal AGV transportation path is generated in combination with a path length punishment mechanism. The algorithm can be prevented from falling into a local optimal solution, and the global exploration capability of the algorithm is enhanced.

Description

Technical Field

[0001] The present invention relates to the technical field of path planning algorithms, and in particular to a path planning method and system of an improved secretary vulture optimization algorithm. Background Art

[0002] Automated Guided Vehicles (AGVs), the core vehicle for intelligent manufacturing and smart logistics, are now used in a wide range of fields, including automotive manufacturing, warehousing and logistics, and the textile industry. Currently, AGV research focuses on two key areas: task scheduling and path planning.

[0003] Traditional path planning algorithms, such as Dijkstra's algorithm, the A* algorithm, the genetic algorithm, and the SBOA algorithm, all suffer from inherent flaws in their computational processes. The Dijkstra algorithm requires traversing all nodes until it finds the endpoint, resulting in slow computation and low efficiency. The A* algorithm's heuristic evaluation function is highly directional, making it prone to missing path branches and becoming trapped in local optima. The genetic algorithm converges too quickly, while the ant colony algorithm converges more slowly and both are prone to becoming trapped in local optima. The SBOA algorithm suffers from uneven initial population distribution, insufficient global search capabilities, and a tendency to become trapped in local optima. In summary, existing algorithms are unable to avoid local optima and enhance global exploration capabilities, lacking robustness and adaptability, resulting in inaccurate path planning results. Summary of the Invention

[0004] The embodiment of the present invention provides a path planning method and system of an improved secretary vulture optimization algorithm, which enhances the accuracy of path planning.

[0005] An embodiment of the present invention provides a path planning method for an improved secretary hawk optimization algorithm, comprising the following steps: obtaining a grid map having a starting point, a target point, and obstacles; an automated guided vehicle using a double chaotic map of a tent map (Tent) and a Bernoulli map (Bernoulli) in an improved multi-strategy algorithm (ISBOA) to randomly generate multiple feasible paths from the starting point to the target point and avoiding obstacles on the grid map as an initial population; using the shortest distance feasible path as a fitness value to construct an optimization target, searching for a global optimal solution in the initial population through a global exploration strategy of the osprey optimization algorithm, and escaping from the local optimum through an adaptive T-distribution interference strategy to complete iterative updating of individuals in the population; and using the fitness value to sort individuals in the iteration results and select the individual corresponding to the minimum fitness value in the sorted results to plan an optimized path.

[0006] Furthermore, the multi-strategy improved algorithm ISBOA specifically comprises the following steps: The double chaotic map of Tent and Bernoulli is used to initialize the population of the secretary bird optimization algorithm SBOA; the global exploration strategy of the osprey optimization algorithm is used to replace the prey search strategy in SBOA; the adaptive T-distribution interference strategy is used to replace the escape strategy of SBOA; and the multi-strategy improved algorithm ISBOA is obtained. Furthermore, the double chaotic mapping of the tent mapping Tent and the Bernoulli mapping Bernoulli is used to initialize the population of the secretary bird optimization algorithm SBOA. The specific steps include: generating a chaotic sequence; for each individual to be generated in the chaotic sequence, generating an initial chaotic value in each dimension of the search space through the tent mapping Tent; inputting the initial chaotic value into the Bernoulli mapping Bernoulli secondary iteration; mapping the chaotic sequence obtained by the secondary iteration to the actual solution space, and connecting the Tent and Bernoulli mappings in series to complete the population initialization.

[0007] Furthermore, the global exploration strategy of the Osprey optimization algorithm is used to replace the strategy of the prey search phase in the SBOA, and the specific steps include: The global exploration strategy formula of the Osprey optimization algorithm in the first stage is: ; in, The fish chosen for the osprey, i For the population i Osprey, j The solution space j dimensions, is a random number between [0,1], The value of is one of {1,2}, is the current position of the Osprey, P 1 means the current position update belongs to the first stage, namely the global exploration stage.

[0008] An embodiment of the present invention provides a path planning system using an improved secretary vulture optimization algorithm, comprising: Path optimization module, used to obtain a grid map with starting points, target points and obstacles; The path planning module is used for automatic guided vehicles. It uses the double chaotic mapping of the tent mapping Tent and the Bernoulli mapping Bernoulli in the multi-strategy improved algorithm ISBOA to randomly generate multiple feasible paths from the starting point to the target point and avoid obstacles on the grid map as the initial population; the optimization target is constructed with the shortest distance feasible path as the fitness value, and the global optimal solution is found in the initial population through the global exploration strategy of the Osprey optimization algorithm. The adaptive T distribution interference strategy is used to jump out of the local optimal solution and complete the iterative update of the population individuals; the fitness value is used to sort the individuals in the iterative results and the individual corresponding to the minimum fitness value in the sorted results is selected to plan the optimized path.

[0009] The embodiments of the present invention provide a path planning method and system using an improved secretary vulture optimization algorithm. Compared with the prior art, the method and system have the following beneficial effects: The multi-strategy improved algorithm ISBOA uses the global exploration strategy of the Osprey optimization algorithm and the adaptive T-distribution interference strategy, so that ISBOA can demonstrate the global exploration ability of the Osprey optimization algorithm and the ability to avoid local optimal solutions of the adaptive T-distribution interference strategy on the planning path of the grid map. Compared with the original SBOA algorithm, it enhances the robustness and adaptability and improves the accuracy of the path planning results. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 An algorithm flow chart provided for an embodiment of the present invention; Figure 2 A 20*20 map comparing the paths of the improved secretary hawk optimization algorithm (ISBOA) and the traditional algorithm provided by an embodiment of the present invention. Figure 3 A path comparison diagram (30*30 map) between the improved secretary hawk optimization algorithm ISBOA and the traditional algorithm provided in an embodiment of the present invention; Figure 4 Iteration curve diagrams of the improved secretary eagle optimization algorithm ISBOA and the traditional algorithm provided in an embodiment of the present invention, where (a) is the iteration curve diagram of a 20*20 map, and (b) is the iteration curve diagram of a 30*30 map. DETAILED DESCRIPTION

[0011] To make the above-mentioned objects, features, and advantages of the present invention more readily apparent, specific embodiments of the present invention are described in detail below with reference to the accompanying drawings. The following description sets forth numerous specific details to facilitate a full understanding of the present invention. However, the present invention can be implemented in many other ways than those described herein, and those skilled in the art may make similar modifications without departing from the scope of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0012] See also Figure 1The embodiment of the present invention provides a path planning method of an improved secretary vulture optimization algorithm, comprising the following steps: Step 1: Construct a multi-strategy improved algorithm ISBOA; wherein, the improvements made by the ISBOA to the secretary bird optimization algorithm SBOA include: using the double chaotic map of Tent mapping and Bernoulli mapping to initialize the population of SBOA; using the global exploration strategy of the osprey optimization algorithm to replace the strategy of the prey search stage in SBOA; using the adaptive T distribution interference strategy to replace the escape strategy of SBOA.

[0013] Step 2: Obtain a grid map with a starting point, a target point, and obstacles. Use ISBOA to plan an optimized path from the starting point to the target point in the grid map while avoiding obstacles.

[0014] Among them, Python is used to randomly generate 20×20 and 30×30 grid maps, NumPy is used to create a two-dimensional array and set the obstacle generation ratio, and a uniformly distributed random number matrix is ​​used for binarization processing. The grids below the threshold are marked as obstacles, and finally Matplotlib is used to realize map visualization, and two sizes of simulation environments are quickly constructed.

[0015] As a new type of swarm intelligence optimization algorithm, the quality of its initial population has a fundamental impact on its performance. The original Secretary Vulture optimization algorithm uses random initialization to obtain the initial population. While this method is relatively simple and effective, and is adopted by most intelligent optimization algorithms, it cannot guarantee a uniform distribution of the initial population across the feasible region, thus affecting search accuracy.

[0016] To solve the above problems, the present invention initializes the population by using Tent combined with Bernoulli's double chaotic map to initialize the population. The basic formula is as follows: .

[0017] .

[0018] in, Represents the new chaotic value after iteration, which is still mapped to the (0,1) interval. represents the chaotic variable value of the nth iteration, with the domain being (0,1), Represents the control parameter, that is, the chaos coefficient, which satisfies , Represents the adjustment coefficient.

[0019] Chaotic mapping has better randomness, regularity and ergodicity, which makes it widely used in optimization algorithms. Initialization through chaotic mapping can increase the diversity of the initialization population, thereby improving the global optimization ability of the algorithm.

[0020] In an optimization iteration simulating the hunting process of a secretary bird, although the original strategy performed well in enhancing algorithm diversity and global search capabilities, as the number of iterations increased, the differential information between individuals gradually decreased, and the algorithm's global search capabilities deteriorated. Therefore, exploring more efficient global exploration strategies became the key to improvement. The Osprey Optimization Algorithm, a swarm intelligence optimization algorithm with strong global exploration capabilities, offers new perspectives and possibilities with its unique global exploration strategy. The Osprey Optimization Algorithm demonstrates excellent global search capabilities by simulating the behavior of an osprey searching for prey in the water. An osprey utilizes its keen perception to search extensively in the water, randomly detecting the location of a food item and attacking it, continuously varying its search direction and depth to find the optimal prey location. We are considering introducing the Osprey Optimization Algorithm's global exploration strategy into the current optimization iteration, replacing the original differential evolution strategy. This replacement is expected to further enhance the algorithm's global search capabilities, enabling individuals to more efficiently explore different regions of the solution space, thereby increasing the chance of finding the global optimal solution.

[0021] The global exploration strategy formula of the Osprey optimization algorithm in the first stage is as follows: .

[0022] in The fish chosen for the osprey, is a random number between [0,1], The value of is one of {1,2}, i For the population i Osprey, j The solution space j dimensions, is the current position of the Osprey, P 1 means the current position update belongs to the first stage, namely the global exploration stage.

[0023] To further enhance the local development capabilities of the SBOA algorithm, this paper introduces an adaptive T-distribution interference strategy during the escape phase of the secretary bird to replace the original SBOA algorithm's escape strategy. This strategy perturbs the population, thereby improving the algorithm's ability to escape local optimality. The adaptive T-distribution strategy is a perturbation strategy that combines global and local exploration capabilities. In the early stages of the iteration, the adaptive T-distribution perturbation is similar to the Cauchy mutation, which gives the SBOA algorithm a stronger global exploration capability. In the later stages of the iteration, the T-distribution perturbation is similar to the Gaussian mutation, which makes the algorithm's local development capability more prominent and also improves its convergence speed. The new position update formula is as follows:

[0024] .

[0025] in, represents the potential better solution generated by the perturbation, represents the potential better solution generated by the perturbation, It represents the number of iterations as the degree of freedom parameter t Distribution function.

[0026] Step 1: If Figure 2 , establish a grid map, specify the starting point and the destination point, and set a number of randomly distributed obstacles on the map; in the map, the black area represents the area where the AGV cannot operate, including each station machine tool, wall, stacking obstacles, etc. The white area represents the area where the AGV can operate.

[0027] Step 2: Set the basic algorithm parameters, population size, number of iterations, etc. The initial population is generated using the double chaotic mapping strategy used by the ISBOA algorithm.

[0028] Step 3: Since the ISBOA algorithm is often used for optimization on continuous domains, it is impossible to directly sample in the grid map used in the present invention. Therefore, all initial populations must be decoded and the continuous codes of the initial populations must be converted into grid numbers. Then, adjacent grid numbers will be selected to determine whether they are continuous.

[0029] Convert the continuous code of the initial population into grid numbers, as shown in the formula: +1 .

[0030] in: is the corresponding grid number, is the corresponding coding position number, is the difference between the starting and ending rows. is the ceiling function, The first consecutive number in the ISBOA digits, is the upper bound of the population encoding for the ISBOA algorithm, For encoding The number of passable grids in the corresponding row is determined by calculating the distance D between the two grids. If D is less than or equal to the distance between the two grids, the two grids are continuous, otherwise they are discontinuous. For discontinuous grids, new grid numbers are generated by continuously taking the midpoints. The new grid numbers are then judged. If the newly added grid is a passable grid, the adjacent grid is replaced. If it is an obstacle grid, the adjacent passable grid is replaced. If all adjacent grids are obstacles, the path is abandoned and a new path is generated. Through the above operations, it becomes a complete and continuous feasible path.

[0031] Step 4: Calculate the fitness of the generated feasible path. The fitness calculation function is as follows: .

[0032] in, Indicates the total distance from the starting node to the current node. and Respectively represent the horizontal and vertical coordinates of the current node in the grid, and Respectively represent the horizontal and vertical coordinates of the next node in the grid, i represents a node in the path, n Indicates the total number of nodes in the path.

[0033] Step 5: The algorithm iteration begins, and the individuals in the population are sorted according to their fitness. The optimization target is constructed with the shortest feasible path as the fitness value. Then, the individuals in the population are updated using the corresponding ISBOA algorithm position update formula. After the individual is updated, the decoding operation is continued, and then the fitness calculation is performed. If the individual position is valid and better than the original individual position, the position is updated, otherwise the original position is retained.

[0034] Step 6: Add 1 to the number of iterations and determine whether it is less than the number of iterations. If it is less, return to step 5 and continue iterating. If it is greater, go to step 7.

[0035] Step 7: After the algorithm iteration is completed, individuals are sorted according to the final fitness, and the global optimal path and optimal solution are obtained.

[0036] The simulation results are shown in Table 1, Table 2 and Figure 2 、 Figure 3 and Figure 4As shown. Comparing it with the SBOA algorithm, OOA algorithm, PKO algorithm and SSA algorithm, it can be seen that the ISBOA algorithm demonstrates excellent performance in the 20*20 and 30*30 grid map environment simulation experiments, and achieves the best results in both optimal path length and average path length, indicating that the algorithm has significant advantages in path optimization. Although the algorithm running time is not the shortest, considering its excellent optimization results, this time cost is reasonable. In addition, the ISBOA algorithm is more likely to escape from the local optimum in complex environments and find the optimal solution in fewer iterations. Therefore, the ISBOA algorithm is an effective path optimization algorithm and is suitable for problems requiring high path optimization accuracy.

[0037] Table 1 Comparison of simulation results for 20*20 maps Table 2 Comparison of simulation results for 30*30 map An embodiment of the present invention provides a path planning system using an improved secretary vulture optimization algorithm, comprising: The path optimization module is used to obtain a grid map with a starting point, a target point, and obstacles.

[0038] The path planning module is used for automatic guided vehicles. It uses the double chaotic mapping of the tent mapping Tent and the Bernoulli mapping Bernoulli in the multi-strategy improved algorithm ISBOA to randomly generate multiple feasible paths from the starting point to the target point and avoid obstacles on the grid map as the initial population; the optimization target is constructed with the shortest distance feasible path as the fitness value, and the global optimal solution is found in the initial population through the global exploration strategy of the Osprey optimization algorithm. The adaptive T distribution interference strategy is used to jump out of the local optimal solution and complete the iterative update of the population individuals; the fitness value is used to sort the individuals in the iterative results and the individual corresponding to the minimum fitness value in the sorted results is selected to plan the optimized path.

[0039] A specific embodiment is as follows: This embodiment discloses a path planning method using an improved secretary vulture optimization algorithm, and the specific steps are as follows: S1. Create a grid map, specify the starting point and target point, and set a number of randomly distributed obstacles in the map.

[0040] S2. Use Tent combined with Bernoulli's double chaotic map to initialize the population, so that the chaotic value distribution is more uniform and has better ergodicity, thereby improving the diversity of the population.

[0041] The initialization process involves setting parameters and generating a chaotic sequence. For each individual to be generated, the initial chaotic value is generated using the Tent chaotic map in each dimension of the search space. Secondly, the chaotic value output by the Tent map is used as input and fed into the Bernoulli map for a second iteration. This uses the Bernoulli map's nonlinear abrupt jump characteristics to break any periodicity in the Tent map. Finally, the mixed chaotic sequence is mapped to the actual solution space. The tandem connection of the Tent and Bernoulli maps effectively combines their ergodicity and nonlinear perturbation capabilities, providing a high-quality initialization population for the optimization algorithm.

[0042] S3. Introduce the global exploration strategy of the Osprey optimization algorithm into the current optimization iteration to replace the original prey search phase strategy of the Secretary Vulture optimization algorithm, further improving the algorithm's global search capability.

[0043] S4. In the escape phase of secretary bird, an adaptive T-distribution interference strategy is introduced to replace the escape strategy of the original SBOA algorithm to improve the algorithm's ability to escape from the local optimum.

[0044] The summary of the present invention is as follows: To address the challenges of the traditional Secretary Eagle Optimization Algorithm (SBOA) in path planning, such as uneven initial population distribution, insufficient global search capabilities, and a tendency to fall into local optima, leading to unstable path quality and inefficient path planning for automated guided vehicle (AGV) transport within a workshop, a multi-strategy improved algorithm (ISBOA) is proposed that integrates dual chaotic mapping, an Osprey search mechanism, and an adaptive T-distribution. This algorithm first generates a highly uniform initial population using the Tent and Bernoulli dual chaotic mappings to enhance global exploration capabilities. It then incorporates the global exploration strategy of the Osprey optimization algorithm to enhance path search robustness in complex obstacle environments. Furthermore, an adaptive T-distribution perturbation strategy is employed to dynamically balance algorithm development and exploration capabilities, effectively avoiding local optima. After constructing a workshop environment model based on a grid method, the continuous domain encoding is converted into a discrete grid path, and an optimal AGV transport path is generated by incorporating a path length penalty mechanism. Comparative experiments on 20×20 and 30×30 grid maps demonstrate that the ISBOA algorithm significantly outperforms algorithms such as the original SBOA and Osprey Optimization (OOA) in terms of optimal path length, convergence speed, and stability, achieving an average path length optimization margin of 18.22%-80.86%. This algorithm effectively avoids local optimal solutions through multiple strategies and significantly enhances global exploration capabilities, demonstrating greater robustness and adaptability, providing an efficient solution for AGV path planning in workshops.

[0045] The above-described embodiments merely illustrate several implementations of the present invention, and while their descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the patent. It should be noted that a person skilled in the art would be able to make numerous variations and improvements without departing from the spirit of the present invention, all of which fall within the scope of protection of the present invention. Therefore, the scope of protection of the patent for this invention shall be determined by the appended claims.

Claims

1. A path planning method based on an improved secretary vulture optimization algorithm, characterized in that: The following steps are involved: Get a grid map with starting point, target point and obstacles; The automated guided vehicle uses the double chaotic mapping of the tent map Tent and the Bernoulli map Bernoulli in the multi-strategy improved algorithm ISBOA to randomly generate multiple feasible paths from the starting point to the target point and avoid obstacles on the grid map as the initial population; the optimization target is constructed with the shortest distance feasible path as the fitness value, and the global optimal solution is found in the initial population through the global exploration strategy of the Osprey optimization algorithm. The adaptive T-distribution interference strategy is used to escape the local optimal solution and complete the iterative update of the population individuals; the fitness value is used to sort the individuals in the iterative results and the individual corresponding to the minimum fitness value in the sorted results is selected to plan the optimized path.

2. The path planning method of the improved secretary vulture optimization algorithm according to claim 1, wherein: The multi-strategy improved algorithm ISBOA specifically includes the following steps: Use the double chaotic map of Tent and Bernoulli to initialize the population of the secretary bird optimization algorithm SBOA; The global exploration strategy of the Osprey optimization algorithm is used to replace the prey search strategy in SBOA. Use the adaptive T-distribution interference strategy to replace the SBOA's escape strategy; Obtain the multi-strategy improved algorithm ISBOA.

3. The path planning method of the improved secretary vulture optimization algorithm according to claim 2, characterized in that: The double chaotic mapping of the tent map Tent and the Bernoulli map Bernoulli is used to initialize the population of the secretary bird optimization algorithm SBOA, and the specific steps include: Generate chaotic sequences; For each individual to be generated in the chaotic sequence, the initial chaotic value is generated through the tent mapping Tent in each dimension of the search space; Input the initial chaotic value into the Bernoulli mapping Bernoulli quadratic iteration; The chaotic sequence obtained by the second iteration is mapped to the actual solution space, and the Tent and Bernoulli maps are connected in series to complete the population initialization.

4. The path planning method of the improved secretary vulture optimization algorithm according to claim 2, wherein: The global exploration strategy of the Osprey optimization algorithm is used to replace the strategy of the prey search phase in the SBOA. The specific steps include: The global exploration strategy formula of the Osprey optimization algorithm in the first stage is: ； in, The fish chosen for the osprey, i For the population i Osprey, j The solution space j dimensions, is a random number between [0,1], The value of is one of {1,2}, is the current position of the Osprey, P 1 means the current position update belongs to the first stage, namely the global exploration stage.

5. An improved path planning system based on secretary vulture optimization algorithm, characterized in that: include: Path optimization module, used to obtain a grid map with starting points, target points and obstacles; The path planning module is used for automatic guided vehicles. It uses the double chaotic mapping of the tent mapping Tent and the Bernoulli mapping Bernoulli in the multi-strategy improved algorithm ISBOA to randomly generate multiple feasible paths from the starting point to the target point and avoid obstacles on the grid map as the initial population; the optimization target is constructed with the shortest distance feasible path as the fitness value, and the global optimal solution is found in the initial population through the global exploration strategy of the Osprey optimization algorithm. The adaptive T distribution interference strategy is used to jump out of the local optimal solution and complete the iterative update of the population individuals; the fitness value is used to sort the individuals in the iterative results and the individual corresponding to the minimum fitness value in the sorted results is selected to plan the optimized path.

Images