Observer-based unknown input and measurement noise estimations for a class of nonlinear systems

This paper provides an estimated scheme of state and disturbance for a class of nonlinear systems with unknown inputs and measurement noise. First, an augmented singular system whose states consist of the state vector and measurement noise vector of the original system is constructed. By choosing a special reduced-order gain matrix, a reduced-order observer which is able to eliminate the influence of the unknown disturbance and nonlinear term of the constructed singular system is developed, and the purpose of estimating both original system state and measurement noise is reached. Second, a high-gain sliding mode observer is considered to estimate the exact derivative of output vector, and a kind of unknown input reconstruction method is proposed. Finally, a modified lateral axis model of an L-1011 in cruise flight conditions is used to illustrate the effectiveness of the proposed approach.