Space object (satellite or space-debris) tracking (SOT) has not received much attention in the Information Fusion community, although the first Fusion conference was held in 1998. A special session on SOT was organized only at the Fusion 2012 conference. Research on tracking at Fusion conferences has focused on ground target tracking, sonar tracking, video tracking, tracking in air traffic control systems, ballistic missile tracking, ballistic projectile tracking, automobile tracking on roads, simultaneous tracking and localization, to name a few. Most of these tracking problems use kinematic models (e.g. nearly constant velocity, nearly constant acceleration, and coordinated turn models), which don't use forces. Ballistic missile and ballistic projectile tracking use dynamic models which are governed by forces. In general, the motion of a space object (SO) is influenced by forces due to gravity, atmospheric drag, solar radiation pressure, and thrust. Sophisticated force models can be used for SOT depending on the tracking accuracy requirements. Secondly, the measurement models used in conventional tracking, such as the angle-only, radar (range, azimuth, elevation), electro-optical/infrared, video (perspective transformation) measurement models are much simpler compared with the radar and electro-optical sensor measurement models used for SOT. Light-time correction, aberration, atmospheric refraction correction, etc. are commonly modeled for SOT, whereas these effects are ignored in conventional tracking applications. However, the filtering and multitarget tracking algorithms used in the Information Fusion community are significantly advanced compared to those used in SOT. In this paper, we introduce some non-trivial dynamic and measurement models for SOT which would be beneficial for researchers in the Information Fusion community. We feel that this would advance further research in multitarget tracking algorithms such as the multiple hypothesis tracking and random finite set based multitarget filtering algorithms (e.g. cardinalized probability hypothesis density and multi-Bernoulli filters).
