Class function-based adaptive disturbance observer for uncertain nonlinear systems

A novel design framework of adaptive disturbance observer for a class of uncertain nonlinear systems based on class $ \mathcal {K} $ K functions is proposed in this paper. In the proposed observer, the upper bounds of the disturbance and its derivatives are not required a priori and the inherent adaptive gain can be arbitrarily selected as any class $ \mathcal {K} $ K functions. It is verified that the estimation error converges into a small neighbourhood around zero asymptotically, wherein the size of the above region can be exactly suppressed to be arbitrarily small by tuning parameters. Compared to conventional methods, the proposed observer is of easy implementations with simple structure and less parameters for tuning. Comparison simulation and an illustrating example are conducted to illustrate the superiorities and effectiveness of the proposed observer.