A Study on Nonunear Robust Adaptive Control System Design of Industrial Robotic Manipulator

Abstract

This paper presents new and simple robust adaptive control system for the implementation of real time control of the multi-joint robot manipulator. Based-on the linearized dynamic equation, the adaptive control scheme used a least square identification(LSAC) is designed to perform the online parameter estimation of the feedback controller gains. Next, the nonlinear robust adaptive control system(NRAC) is designed to perform the robust and real-time control of desired trajectory tracking under the existence of the unknown payload variation and inertia parameter uncertainties. In LSAC scheme a recursive least-square identification scheme is used to estimate the feedback gain matrix directly, not to estimate the inertia parameters of manipulator. The feedback gains of the controller are updated and adjointed in each sampling time to achieve the desired control performance. In NRAC the proposed adaptive control algorithm is developed within the framework of adaptive model following control theory using the Lyapunov Direct Method. The performance of the proposed adaptive control algorithm is compared with a similar adaptive control scheme. To demonstrate the effectiveness of proposed control system, simulation is performanced under the existence of error of inertia parameter and unknown payload.