Multi-input, multi-output flight control design using pseudo-control, software rate limiters, and quantitative feedback theory

Quantitative Feedback Theory (QFT) is a powerful frequency-domain approach to the design of feedback control systems to meet time or frequency domain performance criteria in the presence of parametric uncertainty in the plant dynamics. The QFT methodology has been successfully applied to a variety of aerospace flight control system designs. One shortcoming of the QFT design technique lies in the area of multi-input, multi-output (MIMO) applications where redundant control effectors are in evidence, i.e. more control effectors than output variables to be controlled. In the past, QFT applications to such problems have either ignored the redundant controls, i.e. control effectors were paired, one on one, with output variables, or somewhat ad-hoc techniques were applied to utilize the redundant effectors. The result of these approaches has often been conservative designs with larger closed-loop bandwidths than necessary. This, in turn, has led to flight control systems exhibiting actuator rate saturation with concomitant performance penalties. Given that the goal of the QFT design technique is a closed-loop system with minimum bandwidth exhibiting performance and stability robustness in the presence of significant plant uncertainty, the research presented will provide the methodological development of two procedures aimed at achieving these objectives. The first method employs the use of "pseudo-controls" to distribute control signals among redundant control effectors subject to an optimality criterion, and applies plant pre-compensation to reduce control cross-coupling. These are pre-QFT design techniques intended to reduce closed-loop bandwidth. The second method distributes the control signals using simple actuator ganging and attempts to improve closed-loop system performance with a post-QFT implementation of software rate limiters and modified compensators in an altered feedback control structure. The second method is found to have performance advantages as well as reduced order compensators and prefilters.