Combined NN/RISE-based Asymptotic Tracking Control of a 3 DOF Robot Manipulator

Abstract

Robotic control is an interesting subject due to its extensive industrial applications. This paper proposes a new continuous control design for tracking problem of a three degree-of-freedom (DOF) robot manipulator in presence of uncertainties and bounded external disturbances. The proposed method is based on the combination of the recently developed robust integral of the sign of the error (RISE) feedback and neural network (NN) feed forward terms. In this control strategy, a feed forward NN is utilized to compensate uncertainties of the nonlinear system. Furthermore, the RISE feedback control term is used to eliminate the NN approximation error and bounded external disturbances. Typical NN-based controllers generally provide the convergence only with uniformly ultimately bounded (UUB) error due to the NN reconstruction error. However, the proposed method guarantees asymptotic tracking by eliminating this error. Using Lyapunov stability analysis, a semi-global asymptotic tracking of the robot joints is achieved. Besides, a comparative study on the system performance is conducted between the proposed control strategy and an NN-based controller. Simulation results demonstrate that the proposed controller is robust to deal with uncertainties, and verify the effectiveness of the proposed control scheme.