The Effects of Uncertainty Estimation on Dynamic Sensor Tasking
Patrick S. Williams, David B. Spencer, Richard Scott Erwin, Kyle J. DeMars · AIAA/AAS Astrodynamics Specialist Conference · 2012
The tracking of Earth orbiting objects has been a topic of recent concern, due to the increasing number of man-made orbital debris, active, and inactive space objects over the past several decades. While ground and space-based sensors can provide observations of object characteristics (range, azimuth, elevation, etc), the large amount of objects compared to the limited sensors available to track them results in measurements occurring infrequently, leading to long periods of object uncertainty propagation. These potentially long periods of either inability to make observations (due to line-of-sight access) or unavailability of sensors (due to scheduling constraints) necessitates the need to intelligently determine which objects should be observed and which should be ignored at various times, a process known as sensor tasking. Various methods have been suggested for how to accomplish this tasking, many of which revolve around using an object’s state/uncertainty estimates as a utility metric to prioritize objects for observation. Therefore, the method of propagating and updating (when measurements occur) the state/uncertainty estimates through application of a nonlinear lter can become an important variable in the tasking and overall object tracking performance. Recent studies have shown that for nonlinear tracking problems (including satellite orbit/attitude determination), Gaussian sum lters, such as the adaptive entropy-based Gaussian-mixture information synthesis (AEGIS) lter can provide better uncertainty propagation and updating than the more commonly used extended Kalman lter (EKF) or unscented Kalman lter (UKF). The work presented here investigates applications of these lters to a simpli ed multi-object, multi-sensor satellite tracking simulation to show how the more accurate uncertainty estimation of the AEGIS lter can lead to improved tasking decisions and overall better tracking performance.