Path Planning Method Based on Multi-Layer ELM Optimized A*

Abstract

Global path planning is considered as a basic problem of mobile robot. In this paper, an optimized A* algorithm based on multi-layer Extreme Learning Machine (named as ELM-A *) is proposed to solve the problem of path planning. Taking the environment and the path point information as input, the direction of the next search node is predicted quickly by multi-layer ELM. Then the output direction of multi-layer ELM is used to calculate the cost function of A* algorithm. This method can determine accurate search direction by estimating the influence of obstacles. Simulation results show that compared with RD, RA*and Greedy algorithm, the proposed algorithm reduces the number of traversed nodes and shows better time efficiency stability.