Compensation method for temperature error of fiber optical gyroscope based on relevance vector machine

Aiming to improve the bias stability of the fiber optical gyroscope (FOG) in an ambient temperature-change environment, a temperature-compensation method based on the relevance vector machine (RVM) under Bayesian framework is proposed and applied. Compared with other temperature models such as quadratic polynomial regression, neural network, and the support vector machine, the proposed RVM method possesses higher accuracy to explain the temperature dependence of the FOG gyro bias. Experimental results indicate that, with the proposed RVM method, the bias stability of an FOG can be apparently reduced in the whole temperature ranging from -40 degrees C to 60 degrees C. Therefore, the proposed method can effectively improve the adaptability of the FOG in a changing temperature environment. (C) 2016 Optical Society of America