I. Introduction
In the past several decades, robust adaptive control of nonlinear systems has received much attention for various practical application [1]–[6]. To develop the more effective control techniques for the nonlinear system, the handling of unknown external disturbance and system uncertainty need to be explicitly considered in the control design. Since the practical control problem are characterized by widely changing environmental disturbances and uncertainties, the robust tracking control of uncertain nonlinear systems has been well studied in many literatures [7]–[14]. In the most existing research work, the external disturbance and uncertainty has been assumed that their upper boundary is known and the known boundary is utilized to design the tracking control. To completely use the information about the characteristic of disturbances, the disturbance observer can be developed to approximate the system external disturbance and the approximation output of the disturbance observer is used to design the tracking control.