Abstract:The traditional path planning algorithm suffers from a problem where the search complexity is positively correlated with the resolution of the grid map and the complexity of the obstacle distribution. In this paper, we propose a hierarchical path planning algorithm based on key points generation to address this issue. The algorithm utilizes a key points set to transform the global optimization problem into a parallel local optimization problem, which greatly reduces the complexity of the search space. The method mainly consists of parallel local path planning and global path planning based on Dijkstra algorithm. The parallel local path planning includes both offline prior path planning and online dynamic local path planning. The former generates key points of the global grid map using JPS (jump point search) algorithm, and completes offline prior local path planning according to the key points. The latter generates dynamic local path planning based on the start point, end point and the adjacent key points. The offline prior local path and the online dynamic local path planning together construct a global dynamic undirected graph, which enables the completion of global optimal path planning by the Dijkstra algorithm. Our simulation results demonstrate that as the map complexity and planned route length increase, the proposed method shows significantly improved planning efficiency.

A Hierarchical Path Planning Method Based on Key Points Generation