Control and Robustness Analysis for a High-alpha Maneuverable Thrust-Vectoring Fighter Aircraft


Atesoglu O., ÖZGÖREN M. K.

JOURNAL OF GUIDANCE CONTROL AND DYNAMICS, cilt.32, ss.1483-1496, 2009 (SCI İndekslerine Giren Dergi)

  • Cilt numarası: 32 Konu: 5
  • Basım Tarihi: 2009
  • Doi Numarası: 10.2514/1.42989
  • Dergi Adı: JOURNAL OF GUIDANCE CONTROL AND DYNAMICS
  • Sayfa Sayısı: ss.1483-1496

Özet

This study focuses on designing a nonlinear controller for high-alpha maneuvers of a fighter aircraft with a thrust-vectoring control ability and checking the robustness of the designed controller using the structured singular-value mu-based robustness analysis. The controller is designed using the nonlinear dynamic inversion method. It is designed, to engage either the aerodynamic control surfaces or the thrust-vectoring control paddles of the engines, depending on the flight conditions. The necessary mathematical models are built to describe the nonlinear flight dynamics, the nonlinear aerodynamics, the engine with thrust-vectoring paddles, and the aircraft sensors. The robustness analysis is especially needed when thrust-vectoring control is engaged in a challenging high-alpha maneuver. This is necessary to analyze the effect of increasing uncertainty in the aerodynamic parameters in such a flight condition. In a flight with thrust-vectoring control, the effect of the aerodynamic uncertainties on the robustness is investigated for two different cases. In the first case, the aerodynamic forces and moments are treated as if they are completely unknown. This unusual uncertainty assumption is proposed and investigated for the first time in this paper. In the second case, the aerodynamic forces and moments are assumed to be known but only with a limited degree of confidence (e.g., 70%). The results of the robustness analysis for each case show that it is impossible to achieve a satisfactory robustness if the aerodynamic forces and moments are treated as completely unknown disturbances, whereas robustness can be achieved rather easily if they are known even with only 70% confidence. These conclusions are also verified with numerical flight simulations.