Multi-objective mobile robot path planning algorithm based on adaptive genetic algorithm

Nowadays, mobile robots are more and more widely used in some fields, such as domestic, industry, service. Global path planning is an important technology of robot autonomous navigation. It is related to running speed, running smoothness and safety of robots. In order to solve the problem of insufficient security distance provided by most algorithms, a multi-objective path planning algorithm based on adaptive genetic algorithm that considers path length, path smoothness and safety simultaneously is proposed in this paper. At first, construct navigation maps on the basis of working environments with artificial potential field. Then, several navigation paths are initialized based on the complexity of scenarios. At last, the paths are optimized by the adaptive genetic algorithm, and the optimal path considering length, smoothness and safety is obtained in the light of the Pareto frontier method. In order to achieve the optimal purpose, adaptive operator and supervisory operator were proposed to add or delete path nodes adaptively according to the map complexity. Finally, the performance of the proposed algorithm is verified based on two realistic maps with different complexity. Experimental results show that the proposed algorithm can obtain secure path than other methods, while simultaneously considering length and smoothness.