Intuitive Kinematic Control of a Robot Arm via Human Motion

Abstract

Intuitive kinematic control of a redundant robot arm to track a specified end-effector trajectory is achieved via haptic devices or homogenous robots with the same degrees of freedom (DOFs) and similar mechanical structures. To improve intuitive control, this work proposes a novel method to control directly by human motion the kinematics of a redundant robot arm such that the robot is teleoperated to avoid obstacles and track a given end-effector trajectory using a gesture resembling as human's. This work proposes a kinematic control with an augmented multi-tasking method to solve the local-solution and divergence problems of previous multi-tasking methods when human motion is applied. Computer simulations with a redundant robot arm is used to validate the proposed kinematic-control method. The proposed method enables a redundant robot arm to track a specified end-effector trajectory using a gesture which closely resembles human motion.