Design of a robust gain scheduled controller for the High Incidence Research Model

Accepted for publication in the AIAA Journal of Guidance, Control, and Dynamics, July 1999.

Design of a robust gain scheduled controller for the High Incidence Research Model

G. Papageorgiou and K. Glover

Abstract

In this paper, a robust gain scheduled controller is designed for the High Incidence Research Model (HIRM) that was used in the design challenge set-up by the Group for Aeronautical Research and Technology in EURope (GARTEUR). To this end, two conceptually different methods for synthesising robust gain scheduled controllers are reviewed: the Lyapunov function approach (single and parameter-dependent) and the linear fractional transformation approach. Both techniques require a linear parameter-varying (LPV) model of the real plant for which the controller is designed. A ``quasi-LPV'' model of the longitudinal dynamics of an aircraft is derived that accurately captures the nonlinear behaviour of the aircraft. The robust gain scheduled controller for the HIRM is designed using the derived ``quasi-LPV'' model, the robust gain scheduling method based on parameter-dependent Lyapunov functions, and H-infinity loop shaping. The parameter-dependent Lyapunov function approach is thought to be more suitable for designing robust gain scheduled controllers for supermaneuverable aircraft than the linear fractional transformation approach. The performance of the designed flight controller is very satisfactory.