Path planning of mobile robot with path rule mining based on GA

Path planning is an essential issue in applications of mobile robots. It aims at finding an optimal path for a mobile robot from a starting node to a target node in an environment with obstacles. Inspired by evolution, Genetic Algorithm(GA) is designed to find the optimal solution through encoding a potential solutions to the specific problem on a simple chromosome-like structure and applying the operators on them. As a successful solution of many optimization problems, GA has been widely used to generate an optimal path and got surprising search results. However, the classical GA is more like “blind” search, and evolving without any prior knowledge, which results in inefficiency of GA. To improve the performance of GA in path planning, a path rule mining based on the modified GA is proposed in this paper to improve the search ability of local space in evolution. In the proposed method, path rules are extracted in select phase and used in mutation phase. The simulation results show that the optimal path can be obtained and the efficiency of genetic algorithm is greatly improved.