Finite frequency observer design for T-S fuzzy systems with unknown inputs: an LMI approach\

This paper investigates the problem of H-infinity filtering for T-S fuzzy systems with unknown inputs. The frequency ranges of these external signals are assumed to be known beforehand and to belong in the low frequency band. The observer is designed in the low frequency domain such that the effects of the unknown inputs are attenuated to a specified level gamma by means of an H-infinity performance norm. By exploiting the generalised Kalman-Yakubovich-Popov (GKYP) lemma and the Lyapunov method, sufficient design conditions are derived in linear matrix inequality (LMI) formulations for both continuous-time and discrete-time T-S fuzzy models. Finally an illustrative example is introduced to prove the effectiveness of the proposed approach.