Metaheuristic Approach to Probabilistic Aircraft Conflict Detection and Resolution Considering Ensemble Prediction Systems

This paper presents a methodology to tackle the problem of strategic aircraft conflict detection and resolution, up to 60 minutes in advance, considering wind and temperature uncertainties. The problem of hundreds of en-route aircraft flying multi-segmented 2D trajectories is considered. The weather uncertainty is retrieved from Ensemble Prediction Systems. The conflict detection is based on ensemble trajectory prediction, and is performed using an efficient grid-based procedure. A metaheuristic approach is developed to solve the conflicts, based on simulated annealing, which generates resolution trajectories by modifying the location of the route waypoints (vectoring), with the aim of lowering the probabilities of the conflicts while also minimising the deviation from the nominal paths. Realistic applications in scenarios with different traffic densities are presented.