Application of reinforcement learning to detect and mitigate airspace loss of separation events

* Successful application of Reinforcement Learning to detect and mitigate potential LOS events - Trained model to find potential LOS events - Demonstrated for Runway 22 Arrival - Runway 13 Departure interaction - RL model dynamically discovers set of actions for the Aircraft of Interest that avoid p...

Full description

Saved in:
Bibliographic Details
Published in2018 Integrated Communications, Navigation, Surveillance Conference (ICNS) pp. 1 - 11
Main Authors Hawley, M., Bharadwaj, R.
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.04.2018
Online AccessGet full text

Cover

More Information
Summary:* Successful application of Reinforcement Learning to detect and mitigate potential LOS events - Trained model to find potential LOS events - Demonstrated for Runway 22 Arrival - Runway 13 Departure interaction - RL model dynamically discovers set of actions for the Aircraft of Interest that avoid potential LOS with an Intruder Aircraft * Future Work - Test model with additional trajectories, scenarios - Investigation of model parameters.
DOI:10.1109/ICNSURV.2018.8384989