Soft Fault Diagnosis and Recovery Method Based on Model Identification in Rotation FOG Inertial Navigation System

Inertial navigation system (INS) is a critical and essential equipment for vehicles, ships, and aircrafts. However, as the soft fault parameters of the INS vary with time, internal device operations and external environmental disturbances, periodic diagnosis and recovery of the soft faults are required to satisfy their accuracy requirements. The deployment of human experts for fault diagnosis and recovery in INS would mean low efficiency and heavy workload, as well as low-speed operations. In this paper, a method for automatic diagnosis and recovery of the soft faults, based on the rotation INS (RINS) error model is proposed. This method is implemented by means of a self-rotation mechanism, driven by a specially designated rotation strategy. On the basis of the attitude and change rate of velocity errors in stationary base navigation, a least squares algorithm is used for optimal soft fault parameter identification. Experimental results from a real dual-axis RINS demonstrate the effectiveness of the method in automatically, accurately, and quickly diagnosing and recovering soft faults, and further improving the accuracy of INS, after recovery.