Accurate Ground Impact Footprints and Probabilistic Maps for Risk Analysis of UAV Missions

Abstract

This paper addresses the generation of accurate ground impact footprints and probabilistic maps for fixed-wing UAVs. Monte Carlo simulations are performed using a 6DOF dynamic model of aircraft accounting for wind conditions and different types of uncertainties. Use of generated probabilistic maps for risk analysis is shown along an example of real flight trajectory. To also address possible online use of ground impact footprints, surrogate models (kriging and neural networks) are developed to reduce computation time. Real flight data are used for model and application purposes.