A Comprehensive Probabilistic Assessment Method of UAS Ground Collision Risk

Abstract

Unmanned Aircraft Systems are widely experienced but using these systems for missions near to populated areas raises new safety challenges. To address these challenges, the European Aviation Safety Agency requires assessing, for a given operational profile, the likelihood of on-ground collision with critical infrastructure or people. Various works use Model Based Safety Assessment to identify the failure contributing to the crash, while some works provide probabilistic estimation methods of an on-ground collision. In these methods the assessment is performed thanks to Monte Carlo simulation known to be time-consuming to estimate the probability of rare events. This paper will thus provide a comprehensive tooled method to estimate the on-ground collision probability and uses Importance Sampling method to tackle Monte Carlo limitations. Through a comparative study based on a drone usecase, this paper provides a demonstration of the benefits of Importance Sampling over Monte Carlo method. Indeed, this method allows a reduction in the number of simulations and thus the time needed to compute probabilities, with a high confidence in the results. The experiments are based on a safety model formalized with the Open AltaRica platform on an ad-hoc simulator to perform both Monte Carlo and important sampling simulations.