An adaptive inverse controller for customized fight control systems for general aviation aircraft is presented. The purpose of the system is to render a general aviation aircraft easier to fly via decoupling its flight control system. Artificial neural networks are used to counteract the errors due to nonlinearities and noise and to adapt to partial control failure, thus allowing the pilot to continue to safely control the aircraft. The system was verified with MATLAB simulations for longitudinal flight. Simulations of changing configurations, payload, and partial control system failures have shown that the controller does rapidly adapt to these changes without a need for pilot response. It has also been demonstrated that the controller code can be generated from the Simulink model that is compatible with existing code in the flight computer on a Raytheon Beech Bonanza F33C fly-by-wire test bed. The longitudinal flight theory was extended to lateral-directional flight and an inverse controller was derived. The potential increased safety and certification issues are discussed.
Simulation and Flight Test Assessment of Safety Benefits and Certification Aspects of Advanced Flight Control Systems
2003
62 pages
Report
No indication
English
Aeronautics , Test Facilities & Equipment , Air Transportation , Flight simulation , Flight control , Safety benefits , Certification , Tests , Simulation , Aircraft safety , General aviation , Artificial neural networks , Inverse controls , Software , Decoupling , Adaptive inverse controller , Advanced flight control , General aviation aircraft , MATLAB simulink
User "Type" Certification for Advanced Flight Control Systems
British Library Conference Proceedings | 1994
|Flight Safety Aspects with Modern Flight Control Systems
British Library Online Contents | 1996
|Flight Safety Aspects with Modern Flight Control Systems
British Library Conference Proceedings | 1996
|