Abstract:

In the context of the space-based passive cooperative multi-target angular data association and initial orbit determination mission, to address the problem of poor orbit determination accuracy due to the high association mis-match rate of the traditional method in the complex environment where sensor clutter and missed detection are taken into account, a method for multi-target data association and multi-target initial orbit determination based on parameter correction and clustering estimation is proposed. First, a dynamics model incorporating the Earth's non-spherical J2 term and the Moon's gravitational perturbation and line-of-sight angle measurement model are developed, a sensor clutter and missed detection model is derived and a simulated data generator is constructed. Second, the data association cost matrix is constructed according to the evaluation metrics of the measurement association, the cost matrix processing method based on parameter correction and clustering estimation is proposed, and the optimal allocation algorithm is applied to find out the association results. Third, the global associations are derived based on the cooperative associations and the time-series associations, and the multi-target orbit velocities are solved with reference to the global associations. Finally, the proposed method is validated by numerical targeting simulation experiments with coordinated observation of 10 targets in the geostationary Earth orbit neighborhood. Simulation results show that the proposed method is effective in resisting the influence of sensor clutter and missed detection, improving the association and multi-target initial orbit determination accuracy, and the relative orbit error is less than 1%.

Key words: space situation awareness, passive detection, data association, cooperative orbit determination, multi-target tracking