Accommodation Rule of Double Faults for Seven Inertial Sensors

The reliability of the systems can be enhanced by fault detection and isolation (FDI) method and fault accommodation. FDI methods have been studied from 1960’s in various areas of engineering problems. As reported in literature such as survey papers [Betta et. al., 2000, Frank, 1990] and books [Chow et. al, 1984, Gertler, 1998], various methods of FDI have been studied and applied to diverse applications. To obtain reliability and to enhance navigation accuracy, INS may use redundant sensors. A lot of studies on FDI for the redundant sensors have been performed so far. There are many papers for FDI such as look-up table and squared error(SE) method[Gilmore et. al, 1972], generalized likelihood test(GLT) method[Daly et. al., 1979] and optimal parity test(OPT) method[Jin et.al., 1999] for hardware redundancy. [Yang et.al., 2006, 2007] suggested an accommodation threshold for single fault and accommodation rules for double faults based on the error covariance of an estimated variable, which is related to the navigation accuracy of INS. The accommodation threshold and rules give decision rules to determine whether a faulty sensor should be excluded or not. [Yang et.al., 2007] suggested accommodation rules for double faults when 6 sensors are used. When 6 sensors are used, double faults can be detected, but the faults cannot be isolated in some cases. This paper suggests accommodation rules for double faults when 7 inertial sensors are used, where any double faults can be detected and isolated