Probabilistic Data Association Method for Space Object Tracking

In the optical observation of space objects, multiple measurements often occur in the tracking gate, which brings about the uncertainty of tracking measurement and the reduction of tracking accuracy, causes the instability along the tracking path, and eventually leads to the interruption of tracking and the loss of the target. A new approach, combining the Kalman filter and probabilistic data association, is proposed in this paper for the adaptive tracking of space objects. In this method, the gate of association is predicted by the Kalman filter, while the equivalent measurement obtained from the probabilistic data association is adopted as an effective feed. The experiments show that this technique can effectively improve the tracking accuracy as well as the robustness for the automatic tracking of space objects.