A recently developed adaptive controller based on a disturbance attenuation
problem is applied here to the longitudinal dynamics of the F-18 high angl
e-of-attack research vehicle. For commanded change in angle of attack, the
controller blends aerodynamic controls with thrust vectoring. The blending
is based on the uncertainty of the aerodynamic control moment coefficient v
s the restriction on thrust vectoring using paddles. The development of a c
ontroller for this complex control problem is based an an adaptive controll
er determined from solving, without any approximation, the disturbance atte
nuation problem, where the control coefficient matrix is uncertain. This co
ntroller estimates online not only the system state but also the uncertain
parameters in the control coefficient matrix. The control is determined by
maximizing a nonconcave function with respect to the uncertain parameters a
nd is a function of the disturbance attenuation bound. In linear simulation
s, it is shown that as the bound is decreased, the transient response is si
gnificantly quickened, the thrust vectoring is used more heavily during the
transient, and the steady state is achieved with only aerodynamic control.