Angles-Only Relative Navigation for Target Orbit Refinement with GPS-aided Observer Positioning: Insights from the AstroTwin Project

This paper presents the onboard algorithms developed for the AstroTwin project, designed to perform target orbit refinement and tracking using angles-only relative navigation. The methodology is divided into two distinct modules. The first segment employs a specialized sequential filter as the core algorithm for state estimation, designed to predict the next visibility window and ensure continuous tracking. In the second module, angular measurements are extracted from images captured by the sensor without performing astrometric reduction. The target selection process is described, and one-week simulations in a high-fidelity environment confirm that the filter reduces uncertainty compared to typical TLE estimates.