Multiple-Neural-Networks-Based Adaptive Control for Bilateral Teleoperation Systems with Time-Varying Delays

Abstract

In practical applications, the actual models of the robots in teleoperation systems are difficult to be obtained. In addition, the data-communication exhibit asymmetric variable time delays. All of those will definitely affect the performance of the teleoperation system. In this paper, we use multiple neural networks to approximate the uncertain parts of the system, so as to realize the compensation for uncertain terms in the control scheme. By choosing an appropriate Lyapunov functional, we show that the teleoperation system is stable. The stability criteria of the closed-loop teleoperation system are given in terms of linear matrix inequality (LMI). It is shown that the position errors between the master and the slave, the velocities of each robot converge to zero, asymptotically. Simulation studies which are performed on a nonlinear teleoperation system with 2-two-degree-of-freedoms manipulators are given to illustrate the performance of the proposed control scheme in the presence of asymmetric variable time-delays.