Mobile robot path planning based on dynamic movement primitives

Abstract

A novel learning algorithm based on Dynamic Movement Primitives (DMPs) is proposed for mobile robot path planning. First a path is artificially planned and the trajectories are used as sample set. The autonomous path planning of the robot is realized by establishing the DMPs model, utilizing the model parameters obtained by training with the sample trajectory. At last, the learned trajectory is generalized to new targets to realize the generalization of dynamic movement primitives. The simulation and experimental results show that DMPs algorithm is feasible on a mobile robot path planning.