Geodesic trajectory planning and feedforward controller design for a reconfigurable manipulator

Abstract

To rapidly design and construct a humanoid robot with various configurations with different kinematics and dynamics characteristics, the modular design concept is used in the development of reconfigurable manipulator for the full-size humanoid robot named Robo-Erectus Senior. In this paper, a geodesic trajectory planning method is introduced and applied to the reconfigurable manipulator to achieve the shortest path between two points. Since the dynamics of the modular manipulator will be changed after each reconfiguration, it is crucial to tune the PD control parameters of each modular actuator in order to improve the tracking accuracy and the manipulability of the reconfigurable manipulator. The controllers PD parameters are tuned based on the feedforward plus PD control concept. Finally, the experiments are carried out to verify the proposed approach.