Efficient hybrid Kalman filter for denoising fiber optic gyroscope signal

Abstract

This paper proposes a hybrid Kalman filter algorithm for denoising fiber optic gyroscope (FOG) sensors signal. The algorithms like Discrete Wavelet Transform (DWT), Kalman filter (KF) are successfully applied to denoise the FOG signal in steady state condition. These algorithms fail while denoising the dynamic condition FOG signal. The proposed algorithm efficiently denoises the FOG signal in both static and dynamic conditions. The performance of the algorithm is validated by denoising single axis and three axis FOG with different angular rates. The proposed algorithm is hybridization of KF with adaptive moving average (AMA) technique and named it as adaptive moving average based dual mode Kalman filter (AMADMKF). The AMA is used to detect the discontinuities which select proper Kalman filter gain parameter at different conditions of the signal. The performance of proposed algorithm is compared with Discrete Wavelet Transform (DWT), Kalman filter (KF) algorithms for denoising the FOG signal in both static and dynamic conditions. Allan variance analysis on static signal shows that the amplitudes of rate random walk noise, angle random walk and other noises are reduced by 10 times. The AMADMKF filter applied on static and dynamic data reduces the bias drift by an order of 100. The hybrid KF signal to noise ratio (SNR) improved by 40 dB in case of FOGs static condition. The results show that hybrid KF algorithm is able to reduce the noise of FOG signal in both static and varying angular rate (dynamic) conditions. © 2013 Elsevier GmbH.