Previous Abstract Return to Session F3a

Session F3a: Lunar Positioning, Navigation, and Timing

Advancing Autonomous Navigation: Near-Moon GNSS-Based Orbit Determination
Oliviero Vouch, Andrea Nardin, Alex Minetto, Simone Zocca, Fabio Dovis, Department of Electronics and Telecommunications (DET), Politecnico di Torino; Lauren Konitzer, Joel J.K. Parker, Benjamin Ashman, Goddard Space Flight Center (GSFC) National Aeronautics and Space Administration (NASA); Fabio Bernardi, Simone Tedesco, Samuele Fantinato, Qascom s.r.l.; Claudia Facchinetti, Italian Space Agency
Alternate Number 2

Peer Reviewed

Global Navigation Satellite Systems (GNSSs) have settled as a crucial asset for Positioning, Navigation and Timing (PNT) within the Space Service Volume (SSV), and this technology is increasingly recognized a major player to serve the realm of lunar exploration missions. Current space operations are heavily relying on ground infrastructures, with escalating operational costs and limited resources. Therefore, it is urgent to enhance autonomy of space users, particularly in the task of real-time Orbit Determination (OD). This study aims to demonstrate the performance of GNSS-based onboard OD in the lunar regime. In a sequential Bayesian architecture, where GNSS observations are filtered with an orbital propagator, the sigma-point Unscented Kalman Filter (UKF) model is compared against the renowned Extended Kalman Filter (EKF)-based Orbital Filter (OF). The upcoming LuGRE mission serves as a case study, showcasing near-Moon PNT from a simulated portion of lunar ignition orbit at approximately 61 Earth Radii (RE). Both navigation algorithms are assessed with actual receiver observables, retrieved from a high-fidelity Hardware-in-the-Loop (HIL) simulation. Results highlight that the UKF effectively smooths out harmful Dilution of Precision (DOP) leaps induced by losses of lock of some GNSS signals, while maintaining position estimation errors within 2 km for 98.97% of the time. Moreover, remarkable accuracy gains over the EKF are observed, with a 3? percentile improvement of 79.97% for position estimates and 63.62% for velocity estimates.

Previous Abstract Return to Session F3a