Optimal H_∞ fusion filters for a class of discrete-time intelligent systems with time delays and missing measurement

This paper is concerned with the problem of multi-sensor optimal H_∞ fusion filtering for a class of discrete-time stochastic intelligent systems with missing measurements and time delays. This discrete-time intelligent system model, which is composed of a linear dynamic system and a bounded static nonlinear operator, presents a unified description of delayed or non-delayed intelligent systems composed of neural networks and Takagi and Sugeno (T-S) fuzzy models, Lur'e systems, and linear systems. The missing measurements from multi-sensors are described by a binary switching sequence that obeys a conditional probability distribution. We aim to design both centralized and distributed fusion filters such that, for all possible missing observations, the fusion error is globally asymptotically stable in the mean square, and the prescribed H_∞ performance constraint is satisfied. By employing the Lyapunov-Krasovskii functional method with the stochastic analysis approach, several delay-independent criteria, which are in the form of linear matrix inequalities (LMIs), are established to ensure the existence of the desired multi-sensor H_∞ fusion filters. An optimization problem is subsequently formulated by optimizing the H_∞ filtering performances, which is described as the eigenvalue problem (EVP). Finally, simulation examples are provided to illustrate the design procedure and expected performance.