A linear parameter-varying (LPV) control approach is examined to determine if it is practical to use for largeenvelope e ight control designs. The basic approach is combined with π analysis to reduce conservatism, and the resulting mixed LPV/ π approach is used to design a longitudinal LPV controller for the F-16 Variable Stability In-Flight Simulator Test Aircraft. Implementation issues are discussed, and the controller is modie ed accordingly. The e nal closed-loop system is tested for robust stability and handling qualities throughout the subsonic e ight envelope. High-e delity nonlinear simulations are also used to verify performance. I. Introduction T HE challenge of most control problems is to obtain as much performance as possible in the presence ofplant variations and uncertainty. Unfortunately, robustness often must be sacrie ced to gain performance and vice versa. In some problems adequate levels of performance and robustness cannot be achieved by a e xed controller. This problem occursin e ight control because aircraft dynamicsvary signie cantlyduring operation.To combat this problem, the e ight control designer is forced to design a controller that also varies during operation. The classical approach to developing a varying controller is to design several point controllers throughout the operating region and connect them with some type of blending or interpolation. This process is referred to as gain scheduling (Ref. 1, p. 1) in the literature, and it has several drawbacks. First, designing several point controllersisatime-consumingandtediousprocess.Second,interpolatingorblendingbetweenthepointcontrollersisoftenatrialanderror procedure with historically very little theoretical guidance. Third, anyperformance androbustness guaranteesinthe individualoperatingregionsarelostinthetransitionregionbetweenpointcontrollers. Some modern e ight control methods have eliminated some of the problems with gain scheduling, but not all of them. Dynamic inversion methods (Ref.1,p.25), for example, avoid the scheduling problem by using nonlinear feedback to cancel the dynamics of the aircraft. This enables the controlled dynamics to be specie ed by the designer. Unfortunately, the cancellation is not perfect, and so like classical gain scheduling, dynamic inversion techniques lack solid performance and robustness guarantees. Several new linear parameter-varying (LPV) control approaches 2i 10 have emerged in the last few years, however, that may improve the e ight control designer’ s task. These approaches explicitly take into account the relationship between real-time parameter variations and performance. This enables controllers to be designed for whole ranges of operating conditions with theoretical guarantees of performance and robustness throughout the region. All of the approaches are based (at least in part ) on linear matrix inequalities (LMI), 11 and so they all can be solved numerically with some efeciency.
8
