Ensemble Kalman Filter for Moon Transfer Orbit GPS Navigation

Abstract

This paper addresses the Moon transfer orbit (MTO) GPS navigation and shows the effectiveness of using the ensemble Kalman filter (EnKF) as the navigation filter. The MTO is an orbit going to the moon and we used the one-day orbit to shorten the simulation time. We previously reported the MTO GPS navigation accuracy using the Gaussian filter that was the mixture of the time update of the unscented Kalman filter (UKF) and the observation update of the extended Kalman filter (EKF). The GPS navigation signals coming from the far side of the earth are only available in the MTO and due to both the poor geometry of the visible GPS satellites and the monotone spacecraft movement in the MTO, the navigation suffered from the increase in the error and the position estimation error reached about a few hundred meters at the moon altitude of about 380, 000 kilometers. Because the GPS satellites are given, for the MTO GPS navigation problem, there is only one room left for improving the navigation accuracy: the use of a non-Gaussian filter instead of using the Gaussian filter. The EnKF is one of the non-Gaussian filters for which the states are estimated using non-Gaussian distributions. For nonlinear filtering problems, the superior accuracy of the EnKF over the Gaussian filters has been already confirmed. Therefore, in this paper, we design the algorithm of the EnKF-based GPS navigation filter and investigate its effectiveness for the MTO GPS navigation problem.