Applications of the admissible region to space-based observations

Space-based optical sensors for space situational awareness allow for tracking of objects that are smaller than currently capable with ground-based systems with less influence of atmospheric phenomena. Multiple measurements must be combined in order to obtain a full state description, but their association is uncertain a priori. In this paper, the direct Bayesian admissible region method is applied to space-based observations. Simulation results for both a low-Earth (LEO) and geosynchronous orbit (GEO) observer are shown. For the former, the proposed method successfully associates 101 observations of 8 Earth-orbiting objects taken over the span of 48 h. For the latter, lack of relative motion is identified as a key difficulty in the association process. This problem can be remediated by improved observation strategies.