Neural Network-Based Sensor Online Fault Diagnosis and Reconfiguration for Flight Control Systems

Abstract

A scheme for sensor online fault diagnosis and reconfiguration was proposed. It was based on the radial basis function network (RBF) which was designed by efficient algorithm of on-line training and parameter optimization. Using multiple model adaptive technique, a set of adaptive neural network observers were designed to restrain modeling uncertainties and the output couple in flight control system. The performance of the scheme was validated by the nonlinear simulation for a fighter within automatic terrain following flight control system. As a conclusion, online accommodation is achieved