Optimal flight control of an adaptive aircraft wing modeled by NeuroFuzzy techniques

This work presents a hybrid flight control methodology for adaptive, or morphing, aircraft. Morphing aircraft lack discrete control surfaces and use distributed actuation of the wing surface for maneuvering. Linear Quadratic Regulator (LQR) control is used to determine the optimal aerodynamic forces required to track an arbitrary flight path. The command forces are fed into a fuzzy-logic controller that produces control surface deflections in accordance with preset design rules devised from the control authority of the morphing deflections. Aerodynamic loading is realized from the command control deflections via a neural network model of the vehicle aerodynamics. This work demonstrates accurate tracking of arbitrary flight paths using a hybrid, LQR-NeuroFuzzy control scheme. In addition, we have demonstrated the ability to use multilayer perceptron networks to model steady aerodynamic loads on a morphing aircraft. Finally, this work reveals the capability of using fuzzy-logic to realize desired control methodologies for morphing aircraft based on theoretical insight to aerodynamic performance.