Robust adaptive controller design for telerobots

A controller for telerobots should satisfy definite conditions resulting from specificities of telerobotic operations in unstructured environments. Limitations on on-board computer power and dynamic uncertainties involved in the control process require a simple and robust control law suitable for micro-processor implementation. This law, however, should be able to compensate for large and even potentially unbounded uncertainties. The stability of the control laws becomes the main issue in telerobotics. The known robotic control methods based on feedback linearization or parameter linearization are not applicable to telerobotic control in general. This paper presents a new design procedure for adaptive telerobot controllers. The controller implements a very simple adaptive law (direct approach). In spite of its simplicity the controller demonstrates a total robustness in the sense that the nominal robot dynamic model is not required, and the controller operates with a very general plant model with all terms of uncertain values. The main part of this study is stability analysis in the Lyapunov sense. Simulation results confirm the high performance characteristics of the developed controller.<>