Model-based distributed cooperative control of a robotic manipulator

Abstract

In this paper, a distributed cooperative control scheme is proposed, where each joint-link pair is controlled by a dedicated controller, and the controllers communicate and cooperate among themselves. As the computational load is distributed among the joint level controllers, the computational lead time is reduced in contrast to a centralized model-based control scheme. In addition, unlike in the decentralized (independent joint) control schemes available in the literature, the proposed control architecture does not neglect the coupled dynamics, instead, utilizes the knowledge of the system dynamics completely.