Correlation coefficient stationary series method for gyroscope random drift
Zhihua Wang, Chengrui Liu, FU Hui-min · 2011
Gyroscope plays an important role in navigational systems and its drift has a direct influence on the precision. However, the gyroscope random drift series in practice are not traditional stationary time series for their time-varying means and variances, and large error cannot be avoided by traditional processing method. It is found that the statistical property of most gyroscope random drift series can meet the requirements defined by correlation coefficient stationary series. The analysis method of gyroscope drift series based on correlation coefficient stationary series model is established in this paper. The correlation coefficient stationarity testing method for gyroscope random drift series is discussed firstly. And then the modeling method process is presented. The proposed modeling procedure can efficiently improve the compensating accuracy of gyroscope random drift, because the presented method satisfactorily explains the nonstationary behavior of the gyroscope random drift series commonly existing in practice.