Zonotopic state bounding for 2-D systems with dynamic event-triggered mechanism

Abstract

This paper investigates the problem of zonotope-based state estimation for 2-D systems represented by both the Fornasini–Marchesini local state–space (FMLSS) model and the Roesser model with dynamic event-triggered mechanism (ETM) where the exogenous disturbance and measurement noise are assumed to be unknown-but-bounded (UBB). First, considering the different characteristics of these two models, we design different dynamic ETMs for both FMLSS type and Roesser type respectively. A zonotopic set-membership estimation approach is proposed for the considered 2-D system to obtain the estimation interval. Then, three optimal observer design methods are proposed for different scenarios. In addition, the relationship of the proposed criteria is further discussed between the 2-D augmented error FMLSS system and the 2-D augmented error Roesser system considered in this paper. Finally, simulation results are given to illustrate the effectiveness of the proposed methods.