32th Congress of the International Council of the Aeronautical Sciences

06.1 - Flight Dynamics and Control (Control & Modelling)

FEEDBACK ERROR LEARNING WITH ENHANCED SAMPLING: NEURAL FAULT TOLERANT CONTROLLER OF AIRCRAFT

N. Yoshikawa, Mitsubishi Electric, Japan; S. Suzuki, The university of Tokyo, Germany

Neural adaptive controller trained by feedback error learning enhanced with additional sampling is investigated.rnThe online simulation with the model estimated in real time aids to obtain the additional information.rnThe additional samples are utilized with real samples to train the controller and to improve the convergence of the controller.rn


View Paper