Geostationary Orbit Surveillance Using the Unscented Kalman Filter and the Analytical Orbit Model

A strategy for geostationary orbit (or geostationary earth orbit [GEO]) surveillance based on optical angular observations is presented in this study. For the dynamic model, precise analytical orbit model developed by Lee et al. (1997) is used to improve computation performance and the unscented Kalman filer (UKF) is applied as a real-time filtering method. The UKF is known to perform well under highly nonlinear conditions such as surveillance in this study. The strategy that combines the analytical orbit propagation model and the UKF is tested for various conditions like different level of initial error and different level of measurement noise. The dependencies on observation interval and number of ground station are also tested. The test results shows that the GEO orbit determination based on the UKF and the analytical orbit model can be applied to GEO orbit tracking and surveillance effectively.