On extended state estimation for nonlinear uncertain systems with round-robin protocol

Abstract

This work studies the state estimation problem for nonlinear uncertain systems over a shared communication channel. We transform the original system into an extended state system to handle the nonlinear uncertainty. The round-robin protocol is used to overcome the communication capacity constraint by sending the measurements periodically. The measurements’ structure is changed by the scheduling protocol, which may cause the round-robin-induced periodic extended state system (RRIPESS) to be unobservable. A necessary and sufficient condition of a class of RRIPESS’s observability is established. An extended state estimator is designed. The upper bound of the priori estimation error covariance is derived via a set of periodic Riccati equations (PREs). The stability of these PREs is ensured by the detectability and the stabilizability of the original system’s parameters. In addition, we give an iterative algorithm to improve the upper bound by optimizing a set of scalars induced by Young’s inequality. Finally, a simulation example is given to illustrate the effectiveness of the developed scheme.