A Bias Compensation Method for Ring Laser Gyroscopes Based on Gaussian Process Regression and Distributed Temperature Measurement Configuration

The accuracy of the high-precision ring laser gyroscopes (RLGs) is influenced by various factors in the external multiphysics environment, including temperature, mechanical, and magnetic fields. Among them, complex temperature field changes lead to degradation of gyroscope accuracy, limiting application scope. This article proposes a RLGs bias compensation method based on Gaussian process regression (GPR) and multitemperature point measurement configuration. We conduct a laser gyroscope thermal analysis based on internal structure, construct a multitemperature point measurement configuration that effectively reflects the overall temperature characteristics of the RLG. To solve the nonlinear multidimensional parameter regression problem in the temperature compensation model, we introduce the GPR model for temperature compensation. Validating the proposed method through multiple temperature experiments, the results show it outperforms traditional linear regression models and neural network (NN) models for temperature compensation.