Research of fault detection algorithm based on wavelet analysis in integrated navigation system

Considering the current detection methods are insensitive with soft fault and have large delay problem, by analyzing the basic principle of wavelet singularity being used in fault detection, a algorithm based on modulus maximum principle is proposed in this paper for fault detection of navigation sensors, in this algorithm, only the observation signals of sensors are used to detect fault. The theory of wavelet singularity fault detection and algorithm process are presented in the paper. The algorithm can be used to effectively detect soft faults of the navigation sensors which is difficult to detect and determine the type of faults through the Lipschitz exponent. It can keep the reliability of fault-tolerant integrated navigation system. Simulation results show that the proposed method is effective and practicable.