Robust fault detection for LPV systems using interval observers and zonotopes

In this paper, the problem of robust fault detection using an interval observer for dynamic systems characterized by LPV (Linear Parameter Varying) models is presented. The observer faces the robustness problem using two complementary strategies. Modeling uncertainties are considered unknown but bounded by intervals. Their effect is addressed using an interval state observation method based on zonotope representation of the set of possible states. The observer gain is designed via pole placement using LMI formulation. The method is applied to a LPV representation of a Twin Rotor MIMO System.