An Improved Principal Component Analysis Algorithm on FDI of Redundant Inertial Measurement Unit

In this paper, an improved Principal Component Analysis (PCA) algorithm on fault detection and isolation (FDI) of redundant inertial measurement unit (RIMU) is proposed so as to improve the accuracy of the FDI system. We firstly combine the superiority both on the PCA and parity space method, then take the filter to parity vector in the process. Simulation results show that the accuracy of fault detection has been greatly increased with the addition of the filter, which demonstrates that the proposed method is feasible and can provide a theoretical reference for the FDI system.