Fuzzy gain scheduling for flutter suppression in unmanned aerial vehicles

Abstract

This article describes the creation of a robust fuzzy gain scheduler for flutter suppression in the open-loop response of a non-minimum phase aeroservoelastic UAV (unmanned aerial vehicle) model. Two sets of Takagi-Sugeno (TS) fuzzy rules were constructed for gain scheduling: one set for system identification of the approximate plant matrices and one for full state feedback control using interpolated gains. Interpolation takes place along the one-dimensional, slowly varying velocity envelope. Twenty-three working points, in a velocity range of 20 m/s through 95 m/s, were chosen for the construction of the nominal plant models. Nominal gain vectors were constructed using LQR optimization methods. To achieve stability over the entire velocity envelope, gain vectors were added to the scheduling table using pole placement techniques. The resultant gain scheduling table and fuzzy gain scheduling led to asymptotically stable regulated output responses with average settling times of 0.5 seconds.