Abstract: In view of problems that path planning of the coal mine rescue robot planned by A* algorithm has many path turning points and the path is not smooth enough, a path smoothing algorithm of coal mine rescue robot based on improved A* algorithm was proposed. Firstly, the Douglas-Peucker (D-P) algorithm is used to process the whole path generated by A* algorithm, and eliminate redundant nodes in the path, and extracts several path nodes as key nodes, which solves the problem that there are many redundant nodes and a large number of path turning points of the A* algorithm. Then, the whole path based on the key nodes is fitted by cubic path function, and a smooth path is obtained, which can effectively shorten the path length. The simulation results show that the algorithm has strong universality, although the planning time is slightly increased compared with the A* algorithm, but the planned path turns are few, the path length is short, and the path quality is relatively better than that of the genetic smoothing algorithm.