Hierarchical path planning for autonomous forklifts in complex industrial environments

Abstract

This article aims to quickly plan feasible and collision-free paths for autonomous forklifts operating in complex industrial environments. Inspired by the behaviors of forklifts drivers, a hierarchical path planning approach including two stages is designed, in which a topological map is inferred from the occupancy grid map and a coarse route is generated first, and then a heuristic search method is adopted to solve the planning problem in a short time. This is achieved through novel heuristic functions and exploration strategies under the guidance of the coarse route. Besides, with a trade-off between computational efficiency and accuracy, a multi-disk approximation method for non-convex shapes of vehicles is proposed for fast collision detection. Experiment results verify that the proposed algorithm can generate a feasible path through the narrow passages and meet the real-time requirement.