TETHERED SATELLITE IDENTIFICATION WITH MIXED OBSERVATION DATA

摘要

Tethered satellite systems present a challenge to tracking systems and software, because the end bodies do not follow Keplerian motion. For very long tethers, the motion of the end body can be more consistent with a suborbital vehicle than an orbiting satellite. Therefore, a robust and efficient algorithm is warranted that can identify if a satellite is tethered. This research develops an approach that estimates the tether length by minimizing an objective function of the residual errors between measured and estimated position. Tethered satellite identification is then based on the estimated tether length. A unique feature of the filter is that is permits mixed observations of either end mass, but there is no a priori knowledge of which body is associated with any given data point. This "data sorting" problem is a necessary component of any practical system. The methodology presented here is tested with data from the Tether Physics and Survivability Experiment.