Improved A* algorithm for automatic guided transport vehicle path planning

Abstract

For the automatic guided vehicle (AGV) path planning problem, an improved A* algorithm is proposed for solving it. The algorithm combines the warehouse environment to generate rasterized information. Firstly, an AGV model with a safe driving area is established to avoid the AGV from crossing the obstacle vertices diagonally; secondly, heuristic function is improved in the traditional A* algorithm and an adaptive heuristic function is designed to obtain the globally optimal solution more efficiently; thirdly, a geometric method is used to shave off the redundant linear points and unnecessary corners in the path search; finally, A new feasible point is generated at the corner of the obstacle and a path is obtained that contains a start point, feasible points and an end point. The data output display that the improved A* algorithm can effectively plan a path with shorter distance and safer transportation, reduce the path search time, lower the corner angle of the AGV, and plan an optimal path with higher quality.