An Adaptive Smooth Event-Triggered Approach to Recursive Estimation for Nonlinear Time-Varying Systems With Uncertain Communication Topologies

This article concerns the adaptive smooth event-triggered consistent estimation problem for discrete nonlinear time-varying systems over sensor networks with uncertain communication topologies. The topology switching law obeys a Markovian stochastic process, where the transition probability is not totally known. To save limited communication resources, a novel adaptive smoothing event-triggered mechanism (ASETM) is proposed to manage data transmission in sensor networks that contain smooth-processed historical data and a dynamic triggering threshold. The developed ASETM benefits from smoothing the transmitted data to avoid certain unexpected triggering events, such as system jitter and large random noise. The sufficient conditions for the existence of the estimator are given in terms of recursive matrix inequalities, which guarantee that the filter error system meets both
${H_\infty }$
performance and envelope constraints. Finally, two simulation examples are provided to demonstrate the effectiveness of the developed estimation approach.