State and input estimation for a class of uncertain systems (Q1295070)

The paper considers a class of deterministic systems with a known time-invariant linear part and an unknown nonlinear part which models all uncertain, nonlinear, time-varying terms in the state dynamics. Such a nonlinear part is regarded as an unknown state-dependent time-varying input and the problem of joint state and input estimation based on the observed (linear) output is studied. A suitable equivalent Lyapunov characterization of assumptions commonly used in the literature provides asymptotic estimates to any desired degree of accuracy of the system state and input.