A neural network-based fault detection scheme for satellite attitude control systems

This paper presents an actuator fault detection and identification (FDI) scheme for satellite attitude control systems. A state-space approach is used and a nonlinear-in-parameters neural network (NLPNN) is employed to identify the general unknown fault. The recurrent network configuration is obtained by a combination of feedforward network architectures and dynamical elements in the form of stable filters. The neural network weights are updated based on a modified backpropagation scheme. The stability of the overall fault detection scheme is shown using Lyapunov´s direct method. Simulation results are presented to show the performance of the proposed fault detection scheme