Mode Switching-Based Symmetric Predictive Control Mechanism for Networked Teleoperation Space Robot System

Abstract

Bilateral synchronization control problem for networked teleoperation space robot system (NTSRS) with unknown gravity and asymmetric, random delay as well as communication intermittent interruption is discussed in this paper. To make full use of limited operation window time, improve telepresence, and realize continuous teleoperation, a special mode switching-based symmetric predictive control mechanism is designed, where the mode switching idea is to make full use of limited operation time. Meanwhile, among this mechanism, the symmetric predictive control structure is to improve telepresence and realize continuous teleoperation, besides the same function that mode switching idea provides. The structure of this mechanism for the master side is composed of a predictor and corresponding controller, and symmetric configurations for slave side. Combining with the homomorphic model idea and utilizing the predictive and estimation capability of neural network (NN), novel predictor design method is proposed. In designing the predictor, unlike the traditional model predictive control where predictor is coupled with the controller, this method is independent with the corresponding controller in structure, which is helpful for parameter adjustment and analysis. For unknown gravity, estimation ability of NN is utilized to solve it. With the adaptive method, estimation errors can be kept at a lower level. Subsequently, an analogous proportional-derivative controller with gravity compensation term is designed to realize asymptotical stability and synchronization control of the NTSRS. Finally, effectiveness of the proposed method is verified by some simulation examples.