Dynamic inverse resiliant control for damaged asymmetric aircraft: Modeling and simulation
This paper presents the development and simulation of a dynamic inverse resilient control (DIRC) system that can recover a general aviation aircraft that has sustained damage to a lifting surface in ight, and is now asymmetrical. A new set of di erential equations of motion for the aircraft are derived that account for the asymmetries of the aircraft and for the dynamic e ects of the shift in the center of gravity away from the original body centered axis. The typical symmetric force and moment coe cient buildup is altered to include the inuences of asymmetric lift, drag, and control deection. For the simulation, a Raytheon Bonanaza F-33C y-by-wire testbed is selected since it has been used in both ight and simulation testing for prior Wichita State University research. Geometric, mass, and aerodynamic properties for the damaged aircraft are determined with conceptual design techniques. The DIRC consists of an inverse controller with decoupled ight controls and an adaptive system to correct the command signals for a damaged aircraft. Modeling errors between the inverse control and the actual aircraft are large. Adaptive Bias Correctors, a simpli ed neural network, are applied to adapt the controller to the modeling errors between the undamaged symmetric based inverse controller and damaged asymmetric aircraft. Adaptation happens fast enough to limit the bank angle excursion and bring the aircraft back to wings level ight. Simulations are performed with a MATLAB/Simulink R and a full 6DOF model of the aircraft. Tests show that for losses to the left wing, the DIRC can recover the aircraft until the available control authority is unable to counteract the asymmetric moments.
Thesis (M.S.)--Wichita State University, College of Engineering, Dept. of Aerospace Engineering