Abstract: Path planning is the core technology of mobile manipulators. Aiming at the problem of the traditional fixed-station mobile manipulator planning algorithms, a method for optimizing the picking path of mobile manipulators based on improved genetic algorithm is proposed. By analyzing the position of items to be picked, a sorting path model of mobile manipulators at a single station and a traveling salesman problem(TSP) model for multiple stations are established. An improved genetic algorithm is used to optimize the station coordinates in the workspace, which planned the shortest path grasped by the mobile robot arm and moved between multi-station points. Experiments show that, by using the improved hierarchical evolution selection operator and the optimal nearest neighbor crossover operator, compared with the traditional genetic algorithm, the convergence speed is increased by 46.15%, the path is shortened by 45.99%, and the system running time is reduced by 25.80%. That improved system efficiency.

Key words: logistics sorting, mobile manipulator, path planning, genetic algorithm