Unknown input estimation using reduced-order observers for a class of nonlinear systems

This paper investigates the problem of unknown input estimation for a class of nonlinear systems. The goal of the proposed approach is to estimate the inputs using a reduced-order observer that does not need to estimate the entire state of the considered system. By means of regular transformations and using the constrained generalized Sylvester equation, we exhibit mild conditions for the asymptotic stability. The established sufficient conditions are expressed in term of Linear Matrix Inequality (LMI), which is easily tractable numerically by standard software algorithms. A numerical example is given in order to illustrate the validity of the proposed method.