GNSS-based baseline vector determination for widely separated cooperative satellites using L1-only receivers

Abstract

Real time estimation of the relative position and velocity vectors between two satellites in a formation is an integral part of the formation control loop. Relative positioning based on Global Navigation Satellite Systems (GNSS) has been a dominating technology for formation missions in LEO, where extremely precise estimates could be obtained for formations with small inter-satellite distances (1-10 km). Larger baselines between the satellites (>10 km) are more challenging as they pose the problem of huge differences in the ionospheric delays experienced by the signals received by each receiver. This problem could be mitigated by using precise ionospheric-free combinations that could only be obtained by dual-frequency receivers, which is not a cost-efficient option for the modern low-cost miniature missions. In this paper, the problem of GNSS-based relative navigation between two spacecraft with large inter-satellite distance which are equipped with single-frequency receivers is treated through adopting the space-proven Extended Kalman Filter (EKF). Although using an EKF for relative navigation is a common practice, there are many variants of the filter settings, which vary in terms of the state and measurement vectors to be adopted as well as the techniques to be used to handle the ionospheric delay. In this research, optimal settings of the filter are sought for the problem of relative baseline vector estimation between two spacecraft that have large separation and which are equipped with with single-frequency GNSS receivers.