Cooperative multi-agent model for collision avoidance applied to air traffic management - Archive ouverte HAL Accéder directement au contenu
Article Dans Une Revue Engineering Applications of Artificial Intelligence Année : 2021

Cooperative multi-agent model for collision avoidance applied to air traffic management

Augustin Degas
Elsy Kaddoum
Marie-Pierre Gleizes
Françoise Adreit
  • Fonction : Auteur
Arcady Rantrua
  • Fonction : Auteur

Résumé

In the future, Air Traffic Control (ATC) will have to cope with a radical change in air traffic. Apart from the probable increase in traffic that will push the system to its limits, the insertion of new aerial vehicles such as drones into the airspace-with different flight performances, and utterly different objective-will increase its complexity. Current research aims at increasing the level of automation and/or partial delegation of the control to on-board systems. In this work, we investigate the collision avoidance management problem using a decentralized distributed approach. We propose an autonomous and generic multi-agent system to address this complex problem, where mobile entities are agents, that use a finite set of discrete speed vector modifications to follow their trajectories while avoiding separation losses. We validate our system using a state-of-the-art benchmark. The results underline the adequacy of our local and cooperative approach to efficiently solve the studied problem, competing with centralized methods. Additionally, the conducted sensitivity analysis underlines the robustness of our approach regarding some used parameters.

Dates et versions

hal-03615937 , version 1 (22-03-2022)

Identifiants

Citer

Augustin Degas, Elsy Kaddoum, Marie-Pierre Gleizes, Françoise Adreit, Arcady Rantrua. Cooperative multi-agent model for collision avoidance applied to air traffic management. Engineering Applications of Artificial Intelligence, 2021, 102, ⟨10.1016/j.engappai.2021.104286⟩. ⟨hal-03615937⟩
23 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More