Learning Control Law of Mode Switching for Hypersonic Morphing Aircraft Based on Type-2 TSK Fuzzy Neural Network

Abstract

In this paper, the Linear Parameter Varying (LPV) model and Model Predictive Control (MPC) method are proposed and applied on a morphing wing UAV (MWUAV) for its transient mode. In the model method, an improved function substitution method is introduced , the proposed method combines function substitution and partial linearization, using which the derived LPV model is not necessarily polytope and is more consistent with the original nonlinear model. Then, an MPC controller is derived based on the LPV model. Because the LPV model is polytope, parameter dependent receding-horizon optimization process is introduced and solved with parameter dependent quadratic programming. After that, a comparative simulation with gain-scheduled method based on NSGA-II is performed, the simulation results show that the response of MPC controller can track the control command more accurately with more effective control effort.