基于改进蚁群算法的农业机器人多田块路径规划方法与试验OA北大核心CSTPCD

[Objectives]In response to the impact of fragmented fields and obstacles on agricultural robot operations in hilly areas,this study aimed to propose a multi field block path planning method based on a combination of improved ant colony algorithm and A∗ algorithm.[Methods]By using drones to construct remote sensing images of farmland,using grid method to model the environment of farmland,and then dividing and merging sub regions,a region heuristic function was introduced into the ant colony algorithm to distinguish the movement costs in different regions.The pheromone update method was improved by manually increasing the concentration of pheromones in the optimal path and establishing an adaptive pheromone volatilization coefficient,in order to solve the optimal traversal order of job sub intervals.The A∗ algorithm with heuristic search function was used for sub region connection path planning,ultimately achieving multi field path planning.[Results]The MATLAB simulation test results showed that when the number of sub regions was 40,the average path length planned using the improved ant colony algorithm was reduced by 3.19%compared to the traditional ant colony algorithm,and the average iteration convergence number was reduced by 79.5%.In the path traversal simulation experiment,the coverage rate of the agricultural robot's traversal path could reach 100%,and the path repetition rate was 6.48%.The effectiveness of this method was further verified through simulation experiments of natural environment using remote sensing images of agricultural drones and field operation parameters.[Conclusions]The method proposed in this study could effectively solve the multi field path planning problem of agricultural robots in hilly areas,and provided path reference and technical support for large-scale operation of agricultural robots in hilly areas.