Reliable filtering for switched discrete-time systems with sensor failures: An estimated state-dependent switching method

This paper deals with the problem of reliable H ∞ filtering for a class of switched discrete-time time-delay systems with sensor failures by utilizing the multiple Lyapunov functions method and an estimated state-dependent switching technique. We require neither the stability of each individual subsystem nor the measurability of the system states. A switching rule depending on the filter states is designed, which, together with the filter, can guarantee that switched filtering error systems achieve asymptotic stability and with H ∞ performance γ for all possible sensor failures. In sufficient conditions, we introduce some additional scalar matrix variables to realize the decoupling between the filtering error system matrices and Lyapunov matrices. The reliable H ∞ filter gains can be obtained by solving LMIs. Finally, two illustrative examples are provided to demonstrate the feasibility of the theoretical results.