Fault Diagnosis for Attitude Sensors based on Analytical Redundancy and EMD

This paper proposes a fault diagnosis approach for the satellite attitude determination subsystem based on analytical redundancy and empirical mode decomposition (EMD). Based on the analytical redundancy, a steady-state Kalman filter is proposed to detect the faults in the gyroscopes and the earth sensors. To further locate the fault source, EMD algorithm is used to detect the fault in the gyroscopes. Final decision is based on the results provided these two methods. The proposed method combines the model-based approach and the signal processing technique. Simulation results illustrate the effectiveness of the proposed method.