Maneuver Estimation Model for Relative Orbit Determination

While the use of relative orbit determination has reduced the difficulties inherent in tracking geosynchronous satellites that are in close proximity, the problem is often compounded by stationkeeping operations or unexpected maneuvers. If a maneuver occurs, observations will no longer fit predicted data, increasing the risk of misidentification and cross-tagging. The goal of this research was to develop a model that will estimate the magnitude, direction, and time of a suspected maneuver performed by a collocated satellite in geosynchronous orbit. Relative motion was modelled using Hill\u27s equations, and least squares estimation was employed to create both a linear non-maneuver model and non-linear maneuver model. Two sets of data (DirecTV 4S and AMC-4) for an actual satellite collocation were obtained from the Air Force Maui Optical and Supercomputing (AMOS) site, consisting of differential right ascension and declination. Studies conducted with these observations, along with simulation studies, indicate that it is possible to perform maneuver estimation. It was found, however, that the amount of data required for successful convergence is much greater than that typically obtained for tracking purposes