Drift fault accommodation system of a transport aircraft using Neural Network models

The primary requirement of safety critical systems is to provide acceptable levels of service even in the presence of stuck and drift faults. It is imperative for control systems such as a Flight Control System (FCS) to detect these faults and accommodate the same. This is best done by neural networks (NNs). The work carried out in this paper lies with the induction of drift faults, and designing a Neural Network (NN) to detect and accommodate the sensor fault. The lateral dynamics of the Boeing 747 jet transport aircraft is used to validate the proposed system. All faulty conditions and models are designed and tested in MATLAB/Simulink.