Correlated-Equilibrium-Based Unmanned Aerial Vehicle Conflict Resolution

摘要

Collision avoidance is an essential component of an autonomous robotic system. This paper presents CONCORD, a novel decision-making system for unmanned aerial vehicle (UAV) conflict resolution that uses the notion of correlated equilibrium. Correlated equilibrium is a solution concept in game theory, which is computationally light and gives a better-expected payoff to every player. This paper proposes an intelligent semi-autonomous conflict resolution system for aerial vehicles. The vehicles are the players, and the resolutions are the equilibrium strategies of the game. CONCORD monitors the vehicles to predict conflicts and compute strategies leading to equilibrium. The cases of cooperative, non-cooperative, and multiple vehicles are discussed, emphasizing decision making and its corresponding timelines. Monte Carlo simulations are also performed to demonstrate the effectiveness of the proposed framework. The proposed framework is compared with two other baselines and a Nash-equilibrium-based resolution to highlight its effectiveness. This work also presents the integration of CONCORD to an existing UAV traffic management system. The proposed module would enable conflict-free flight for a multi-UAV system in an efficient manner.