AGV Path Planning Based on Improved Ant Colony Algorithm and Dynamic Window Method
The automated guided vehicle(AGV)is an unmanned material transport device equipped with an electromagnetic or optical automatic guiding device enabling it to travel along a specified path,stop and wait accurately at designated positions,and carry out handling activities as per operation requirements.In the modern logistics system,AGVs have greatly optimized the warehousing and distribution process by improv-ing the efficiency and accuracy of material handling,thereby significantly improving the overall performance of the system.Path planning refers to the practice where an algorithm is used to search for an optimal or sub-optimal path for the AGV to traverse a map without collision,the purpose of which is to efficiently and safe-ly guide the AGVs to move in a complex logistics environment to complete a task smoothly.Common glob-al static path planning algorithms include:ant colony algorithm,A*algorithm,and Dijkstra algorithm,etc.,while common local dynamic path planning algorithms include:dynamic window method,and artificial po-tential field method,etc.In this paper,according to the principle of algorithmic fusion,we proposed the improved ant colony dy-namic window path planning algorithm,where the initial pheromone,heuristic function,and pheromone update rule of the ant colony algorithm are improved and fused with the dynamic window method.The op-timal path obtained by the ant colony algorithm is used as the global guide for the dynamic window method,and the evaluation subfunction of the dynamic window method is added in the process so as to reduce the AGV path length and improve its smoothness.In order to verify the effectiveness of the improved ant colony algorithm,we established the environmen-tal model using the grid mapping method,and had a simulation experiment using Matlab for comparison,finding that due to the improvement in initial pheromone,heuristic function,and pheromone update rules,the final path obtained using the improved ant colony algorithm in this paper was greatly improved over the traditional ant colony algorithm.Despite the small difference in length of path,the improved ant colony algo-rithm was far lower than the traditional algorithm in terms of number of iterations,convergence speed and path-finding smoothness.Finally,in order to verify the effectiveness of the improved ant colony/dynamic window algorithm,we compared the traditional dynamic window algorithm and the fusion algorithm of this paper through Matlab,which found that the fusion of the improved ant colony algorithm gave the dynamic window method a glob-al optimal path guidance,which could accurately find the end point in complex terrain,and keep the AGV as close to the optimal path as possible at all times,thereby shortening path length and improving path smoothness.
AGVimproved ant colony algorithmdynamic window methodpath planning
NETL
NSTL
万方数据
