AIR FORCE INST OF TECH WRIGHT-PATTERSON AFB OH SCHOOL OF ENGINEERING
One of the conclusions from the STOLMTD program was the need for a multivariable method of designing controllers of low order. This research investigated that problem by studying reduced order mixed H-twoH-infinity control theory applied to the STOL Landing configuration which employs both thrust vectoring and the use of a canard. Model matching techniques were used to obtain responses that met handling qualities criteria and reduced pilot workload by decoupling pitch rate and velocity commands. The time responses were found through nonlinear simulation and showed that the full order designs did match the ideal models very well and had good noise and wind rejection. Singular value analysis showed that the commands were decoupled very well. The reduced order method was mixed H-twoH-infinity optimization. A fourth order controller that had good performance was found by using a performance constraint, and a fourth order controller that provided good margins was found using a robustness constraint. A third order controller was also found with a performance constraint. Recommendations for finding a low order controller, with good performance and robustness are given.