A Continuous Reconstruction Observer for Sampled-Data Linear Time Varying Systems

Abstract

In this paper, a novel continuous sampled-data observer for linear time-varying systems is presented. The proposed observer is based on two different impulsive observers. The state estimates of these impulsive observers are fused in a manner such that continuous state estimation is achieved. Another significant contribution is comprehensive stability analysis of the proposed observer. The analysis establishes conditions that guarantee exponential convergence of observers. Contrary to the common understanding, it is revealed that the convergence of associated discrete-time equation for impulsive observers is not a sufficient condition for overall convergence. Contributions are illustrated by simulations.