Improved conflict detection and resolution for service UAVs in shared airspace

• Published in: IEEE Transactions on Vehicular Technology
• Main Authors: Ho, Florence, Geraldes, Ruben, Goncalves, Artur, Cavazza, Marc, Prendinger, Helmut
• Format: Article in Journal/Newspaper
• Language: English
• Published: Institute of Electrical and Electronics Engineers (IEEE) 2019
• Subjects: QA75 Electronic computers. Computer science; Orca
• Online Access: http://gala.gre.ac.uk/id/eprint/22512/; http://gala.gre.ac.uk/id/eprint/22512/1/22512%20CAVAZZA_Improved_Conflict_Detection_and_Resolution_2018.pdf; https://doi.org/10.1109/TVT.2018.2889459

In future UAV-based services, UAV (Unmanned Aerial Vehicle) fleets will be managed by several independent flight operation service providers in shared low-altitude airspace. Therefore, Conflict Detection and Resolution (CDR) methods that can solve conflicts---possible collisions between UAVs of different service providers---are a key element of the Unmanned Aircraft System Traffic Management (UTM) system. As our CDR method, we introduce an adaptation of ORCA, which is a state-of-the-art collision avoidance algorithm hitherto mainly used in a limited theoretical scope, to realistic UAV operations. Our approach, called Adapted ORCA, addresses practical considerations that are inherent to the deployment of UAVs in shared airspace, such as navigation inaccuracies, communication overhead, and flight phases. We validate our approach through simulations. First, by empirically tuning the ORCA parameters look-ahead time window and deconfliction distance, we are able to minimize the ORCA generated deviations from the nominal flight path. Second, by simulating realistic UAV traffic for delivery, we can determine a value for separation distance between UAVs that uses airspace efficiently.