Event-Triggered Distributed State Estimation for Nonlinear Reference Systems Under Markovian Switching Topologies

This paper proposes a novel distributed observer for state estimation of a class of nonlinear systems under Markovian switching topologies. The communication constraint is that each observer can not observe the reference state individually, but the network is jointly observable. Since communication energy and bandwidth are limited in engineering practice, an event-triggered communication scheme is developed to decrease the transmission frequency. A sufficient condition is then given to guarantee that the estimation error is exponentially bounded in mean square under the Markovian switching topologies and event-triggered mechanism, which is rigorously analyzed. The effectiveness of the proposed method is verified through the Chua's circuit system.