A WAVELET-BASED METHOD FOR PROCESSING SIGNAL OF FOG IN STRAP-DOWN INERTIAL SYSTEMS

Abstract

Fibre optical gyroscopes (FOGs) have been applied widely in many fields in contrast with their counterparts such as mechanical gyroscopes and ring laser gyroscopes. The precision of FOG is affected significantly by bias drift, angle random walk, temperature effects and noises. Especially, uncertain disturbances resulting from road irregularities often affect accuracy of strap-down inertial system (SINS). Hence, eliminating uncertain disturbances from outputs of a FOG plays a crucial role to improve accuracy of SINS. This paper presents a wavelet-based method for denoising signals of FOGs in SINS used for exploring and rescuing robots in coal mines. Property of road irregularities in mines is taken into account as a key factor resulting in uncertain disturbances in this research. Both frequency band and amplitude of uncertain disturbances are introduced to choose filtering thresholds. Experimental results have demonstrated that the proposed method can efficiently eliminate uncertain disturbances due to road irregularities from outputs of FOGs and improve accuracy of surrogate data. It indicates that the proposed method has a significant potential in FOG-related applications.