Online learning of a motor map for humanoid robot reaching

Abstract

We propose a control system for humanoid robot reaching using a motor-motor mapping that is learnt online. The system combines endpoint closed-loop and open-loop visual servo control. The closed-loop component moves the eyes, head, arm, and torso, based on the position of the target and the robot’s hand, as seen by the robot’s head mounted cameras. The learned, open-loop component brings the hand into view.