The ERP Prediction Method Based on Phase Space Reconstruction Theory and Volterra Series

摘要

The Earth Orientation Parameters (EOP) are necessary to realize the mutual conversion between ICRF and ITRF, and they are of great significance for the orbit determination of high-precision spacecraft. However, when the coordinate transformation is used in the calculation of the autonomous orbit of the satellite, the EOP used can only be obtained through prediction. As important parameters of EOP, polar motion (PM) and length-of-day (LOD) are collectively referred to as Earth Rotation Parameters (ERP). Improving the prediction accuracy of ERP will greatly affect the performance of satellite autonomous navigation. In this paper, given the nonlinear time-varying characteristics of the earth rotation, we used the least square (LS) fitting of the original ERP sequence to get the residual sequence and then used the phase space reconstruction (PSRT) and the Volterra adaptive prediction algorithm to forecast. The experiment results are compared with the results of the earth orientation parameters prediction comparison campaign, and the comparison results prove that our method has certain reference significance and application value in the ERP forecast work.