Linear-parameter-varying gain-scheduled control of aerospace systems

The dynamics of many aerospace systems vary significantly as a function of flight condition. Robust control provides methods of guaranteeing performance and stability goals across flight conditions. In μ-syntthesis, changes to the dynamical system are primarily treated as uncertainty. This method has been successfully applied to many control problems, and here is applied to flutter control. More recently, two techniques for generating robust gain-scheduled controller have been developed. Linear fractional transformation (LFT) gain-scheduled control is an extension of μ-synthesis in which the plant and controller are explicit functions of parameters measurable in real-time. This LFT gain-scheduled control technique is applied to the Benchmark Active Control Technology (BACT) wing, and compared with μ-synthesis control. Linear parameter-varying (LPV) gain-scheduled control is an extension of H_∞ control to parameter varying systems. LPV gain-scheduled control directly incorporates bounds on the rate of change of the scheduling parameters, and often reduces conservatism inherent in LFT gain-scheduled control. Gain-scheduled LPV control of the BACT wing compares very favorably with the LFT controller.; Gain-scheduled LPV controllers are generated for the lateral-directional and longitudinal axes of the Innovative Control Effectors (ICE) aircraft and implemented in nonlinear simulations and real-time piloted nonlinear simulations. Cooper-Harper and pilot-induced oscillation ratings were obtained for an initial design, a reference aircraft and a redesign. Piloted simulation results for the initial LPV gain-scheduled control of the ICE aircraft are compared with results for a conventional fighter aircraft in discrete pitch and roll angle tracking tasks. The results for the redesigned controller are significantly better than both the previous LPV controller and the conventional aircraft.