Missile target accuracy estimation with importance splitting

Missile safety takes a more and more important place in the design and the evaluation of a missile. For that purpose, missile performance collateral damages can be characterised with rare quantiles around the missile target. Usual methods like Monte Carlo simulations are unfortunately not efficient to estimate rare quantiles. Consequently, we propose to apply an advanced method of rare event estimation called importance splitting to decrease the quantile relative error. We show the performance of this algorithm on a realistic missile case.