The massive growth of space activities over the past decades has lead to an increase of the space debris population, which is threatening the safety of space operations and active satellites. Several fragmentation events caused by catastrophic collisions and explosions have single-handedly contributed to the increase in the number of space debris, highlighting the importance of the reconstruction of such events to gain insight on the breakups and decrease the collision risk they cause to active satellites. This work proposes a fragmentation reconstruction approach for this purpose, which combines pruning and clustering criteria with a backward propagation to identify the epoch of the fragmentation and the involved objects. The model was tested by using publicly available Two-Line-Elements data. The problem of the unreliability of such data is addressed by modifying the original approach to account for uncertainty, multiplying the original Two-Line-Elements data and choosing the optimal set of objects for the reconstruction. The optimal values of the pruning filters thresholds for the method are also investigated to identify the most effective criteria. Moreover, the reciprocal influence of the filters is assessed through a sensitivity analysis. The proposed approach is applied to two real fragmentation events to compare the accuracy of the reconstruction with and without the uncertainty quantification. The method proves to be effective in the evaluation of the epoch of the breakup as well as the identification of the fragments and their parent. • Fragmentation reconstruction provides key insights on breakups and collisions. • Breakups can be rebuilt via pruning and clustering with backward propagation. • Two-Line-Elements uncertainty can be handled through a Gaussian Mixture Model. • The minimum orbital intersection and distance margin filters most affect pruning.
Ottoboni et al. (Tue,) studied this question.