An improved nonlinear extended state observer with adaptive variable gain

Abstract

This paper proposes an improved nonlinear extended observer with adaptive gain (ANESO), which can be applied to systems with large disturbance amplitude changes without parameter re‐tuning. An ANESO is added an adaptive factor Ai$$ {A}_i $$ to the gain of the original NESO. The expression of Ai$$ {A}_i $$ is obtained by assuming that nonlinear extended states observer (NESO) and linear extended states observer (LESO) has the same characteristic of steady‐state error when the amplitude of disturbance turns. On this basis, an adaptive adjustment algorithm based on tracking error is designed. In addition, the stability of the proposed ANESO is analyzed by using the Routh‐Hurwitz criterion. Finally, through simulation experiments, the observation performance of ANESO was compared with NESO and switching extended state observer (SESO). The experimental results demonstrate that the observation performance of ANESO is hardly affected by the disturbance amplitude, and moreover, its observation precision is higher than that of NESO and SESO.