Multiple Hypothesis Tracking (mht) for Space Surveillance: Results and Simulation Studies

The need to accurately track breakups and satellite clusters, as well as to resolve uncorrelated tracks (UCTs), especially uncorrelated optical observations (UCOs), requires new, robust, and autonomous methods for space surveillance to enable the development and maintenance of the space catalog and to support the overall SSA mission. This paper presents results on the performance of a newly-developed, statistically-robust, system-level, multiple hypothesis tracking (MHT) capability for joint catalog maintenance, UCT/UCO resolution, and initial orbit determination, in multiple regimes of space. A recent companion paper described some of the unique components contained within the authors’ implementation of MHT, including the multiframe data association problem and the statistical framework used for scoring the likelihood that a sequence of measurements, UCTs/UCOs, and other reports emanate from a common object (either newly-discovered or already catalogued). As demonstrated by the results in this paper, the MHT system provides superior tracking performance (compared to existing methods for data association) in realistic multi-sensor multiobject tracking scenarios over multiple regimes of space. Specifically, we demonstrate that the prototype MHT system can accurately and efficiently process tens of thousands of UCTs/UCOs emanating from thousands of space objects in realtime on a single processor.