Model Experiments of the Control System of the Underwater Manipulator

Abstract

This paper studies the control system of the underwater manipulator of an underwater vehicle. The learning feed-forward controller is applied to the hybrid position and force control system of the underwater manipulator. This control system consists of a feed-forward controller based on an inverse-dynamics model of the controlled system and a feedback controller like a proportional, integral and derivative controller. This control system is adaptable to the nonlinear dynamics of multi-link manipulator and the change of dynamics both of the manipulator and of the controlled object. For the learning of the inverse-dynamic model a new method is proposed to hasten the learning speed. Experiments carried out showed that the new learning method is quick and that the control system is adaptable to the change of dynamics according to the payload.