A new multi-target tracking algorithm for a large number of orbiting objects

This paper demonstrates the applicability of the filter for Hypothesised and Independent Stochastic Populations (HISP), a multi-target joint detection/tracking algorithm derived from a recent estimation framework for stochastic populations, to wide area surveillance scenarios in the context of Space Situational Awareness. Designed for multi-object estimation problems where the data association between targets and collected observations is moderately ambiguous, the HISP filter has a linear complexity with the number of maintained tracks and the number of observations, and is a scalable filtering solution adapted to large-scale target tracking scenarios. It is illustrated on a challenging surveillance problem involving 30 targets on different orbits, observed by 3 sensors with limited coverage, measurement noise, false alarms, and missed detections.