A dynamically event-triggered approach to recursive filtering with censored measurements and parameter uncertainties

In this paper, a dynamically event-triggered filtering problem is investigated for a class of discrete time-varying systems with censored measurements and parameter uncertainties. The censored measurements under consideration are described by the Tobit measurement model. In order to save the communication energy, a dynamically event-triggered mechanism is utilized to decide whether the measurements should be transmitted to the filter or not. The aim of this paper is to design a robust recursive filter such that the filtering error covariance is minimized in certain sense for all the possible censored measurements, parameter uncertainties as well as the effect induced by the dynamically event-triggered mechanism. By means of the mathematical induction, an upper bound is firstly derived for the filtering error covariance in terms of recursive matrix equations. Then, such an upper bound is minimized by designing the filter gain properly. Furthermore, the boundedness is analyzed for the minimized upper bound of the filtering error covariance. Finally, two numerical simulations are exploited to demonstrate the effectiveness of the proposed filtering algorithm. (C) 2019 The Franklin Institute. Published by Elsevier Ltd. All rights reserved.