The master adaptive impedance control and slave adaptive neural network control in underwater manipulator uncertainty teleoperation

Abstract

This paper proposes a novel bilateral adaptive control scheme to achieve position and force coordination performance of underwater manipulator teleoperation system under model uncertainty and external disturbance. A new nonlinear model reference adaptive impedance controller with bound-gain-forgetting (BGF) composite adaptive law is designed for the master manipulator force tracking of the slave manipulator. The reference position in task space is obtained from the linear second-order impedance model whose input is the force error of the master and the slave. The adaptive terminal sliding mode control based on adaptive uncertainty compensation is proposed to achieve the master position tracking of the reference position. The radial basis function neural network (RBFNN) local approximation method is proposed for the slave manipulator's position tracking. The RBFNN based on Ge-Lee (GL) matrix is adopted to directly approximate each element of the slave manipulator dynamic, and the robust term with a proper update law is designed to suppress the error between the estimate model and the real model, and the external disturbance. The asymptotic tracking performance and global stability of the teleoperation system are proved with Lyapunov stability theorem. The simulation and experiment verify the performance of the proposed controller in teleoperation manipulator model. The results show that the teleoperation system has a good ability of position and force coordination.