Nonlinear empirical mode predictive drift extraction on fiber optical gyroscope

The integration drift of inertial unit is a fatal error on navigation and attitude determination system. Many approaches have been developed to extract and eliminate it. A novel fiber optical gyroscope drift extraction algorithm was proposed in this paper to ameliorate the performance of fiber optical gyroscope by extracting the trend component as the compensation. In this algorithm, the attitude quaternion of stellar sensor and the output of fiber optical gyroscope were imported into a nonlinear empirical mode predictive algorithm to extract the drift component of gyroscope. Combing the advantages of predictive filter and empirical mode decomposition in nonlinear signal processing, our proposal is more precise in drift extraction and data approximation. Software simulation and hardware verifications on gyroscope were launched, in which the results had proven the capability of the algorithm.