Optimization of robot path planning parameters based on genetic algorithm

Yuhong Liang, Fating Hong, Qinjie Lin, Sheng Bi, Liqian Feng

Abstract: The paper introduces a method which is based o n genetic algorithm to select more properly parameters for local path planning of mobile robot. In our experiments, according to experience, we choose properly parameter domain as chromosomes in Genetic algorithm(GA). Then we will extract several different individual cost function from DWA algorithm, adopted by Local path planning, and merge them into a fitness function. Experiments show that the automatic tuning method, proposed in this paper, can generate smoother and faster trajectories than manual tuning method and perform better in avoiding obstacle. This method can also be used for other platforms.