A Machine Learning Approach for Conflict Resolution in Dense Traffic Scenarios with Uncertainties

Duc-Thinh Pham (1) , Ngoc Phu Tran (1) , Sameer Alam (2) , Vu Duong (2) , Daniel Delahaye (3)

1 Air Traffic Management Research Institute
2 Air Traffic Management Research Institute
3 ENAC - Ecole Nationale de l'Aviation Civile

With the continuous growth in the air transportation demand, air traffic controllers will have to handle increased traffic and consequently more potential conflicts. That gives rise to the need for conflict resolution tools that can perform well in high-density traffic scenarios given a noisy environment. Unlike model-based approaches, learning-based or machine learning approaches can take advantage of historical traffic data and flexibly encapsulate the environmental uncertainty. In this study, we propose an artificial intelligent agent that is capable of resolving conflicts, in the presence of traffic and given uncertainties in conflict resolution maneuvers, without the need of prior knowledge about a set of rules mapping from conflict scenarios to expected actions. The conflict resolution task is formulated as a decision-making problem in large and complex action space, which is applicable for employing the reinforcement learning algorithm. Our work includes the development of a learning environment, scenario state representation, reward function, and learning algorithm. As a result, the proposed method, inspired from Deep Q-learning and Deep Deterministic Policy Gradient algorithms, can resolve conflicts, with a success rate of over 81%, in the presence of traffic and varying degrees of uncertainties.

Mots clés: Reinforcement learning; air traffic control; deep deterministic policy gradient; conflict resolution

Domaines: Optimisation et contrôle [math.OC]

Duc-Thinh Pham, Ngoc Phu Tran, Sameer Alam, Vu Duong, Daniel Delahaye. A Machine Learning Approach for Conflict Resolution in Dense Traffic Scenarios with Uncertainties. ATM 2019, 13th USA/Europe Air Traffic Management Research and Development Seminar, Jun 2019, Vienne, Austria. ⟨hal-02138135⟩