Multi Mobile Robot Path Planning Based on Rough-Fine Search Strategy

Abstract

Multi-mobile robot path planning technology is the key technology in the field of intelligent warehousing and intelligent logistics. Solving the optimal solution for multi-mobile robot path planning will cause “dimension explosion”. Existing multi-mobile robot path planning algorithm could not get the solution with high quality fast, causes the low efficiency of multi-mobile robot systems. This paper proposes a multi-mobile robot path planning method based on rough-fine search strategy. Firstly, the environment map is discretized to two kinds of grid maps with different grid sizes. And then, the M* algorithm is used to obtain the rough optimal solution in the grid map with large size, in the next, the rough optimal solution was mapped to the fine grid map as the teaching path, Finally, in the fine grid map, select the Gaussian distribution band with teaching path as the mean value. In the Gaussian distribution band, the RRT* algorithm is used to optimize the teaching path. Through experimental verification, our algorithm can obtain a higher quality path and has a faster solution speed.