A Novel Multilateral Control Design for Delayed Nonlinear Teleoperation System With RBFNN-based Environments

Abstract

Abstract In this paper, a novel multilateral control design for nonlinear teleoperation system is proposed to improve the capability of multiple robots to coordinate efficiently and precisely in the remote environments under time-varying delays and various uncertainties. The environment is modeled with a general form of force under radial basis function neural network(RBFNN)-based identification and reconstruction to avoid the passivity issue in the traditional teleoperation control and provide the human operators with good sensing of environments. The desired trajectory producers and RBFNN-based sliding mode controllers are designed separately to achieve the good tracking of master/slave robots, and the coordinated distribution algorithm is designed to obtain the control input uS,i for each slave robot. Therefore, the global stability, good transparency performance with both position tracking and force feedback, and good cooperative performance can be achieved simultaneously for delayed nonlinear teleoperation system. The real platform experiment is carried out on a 2-master-2-slave teleoperation system to verify the effectiveness of proposed control design.