Real-Time Conflict Resolution Algorithms for Autonomous Drone Traffic Management

Abstract

Autonomous drones are increasingly being used in urban areas, and with an increase in the number of users, the danger of mid-air collisions and congestion has emerged, calling for effective real-time conflict resolution systems. They introduce a new algorithmic framework based on predictive trajectory modeling, dynamic prioritization, and multi-agent coordination to manage drone traffic safely and efficiently in the context of high-density drone traffic. The framework regularly monitors the position of the drones and calculates potential collision situations while modifying the path so as to maintain the minimum safe distances while not hindering the overall use of the airspace. The results of the performance were comprehensively tested through intensive simulations, where the algorithm was tested against a rule-based algorithm and a classical multi-agent approach. The proposed technique is found to be significantly superior to the baseline techniques, with a conflict resolution rate of 94.5% and an average computation time of 18ms compared to 363ms for the baseline techniques, and also, avoiding 96.7% of potential collisions. The outcomes confirm the scalability, real-time applicability, and robustness of the framework in complex urban airspaces. The limitations are based on the assumption of accurate position data and the difficulty that occurs under extreme environmental conditions, indicating directions for future improvements. The research offers a feasible and scalable approach to AADT and serves as a foundation to incorporate AI-based adaptive learning, heterogeneous fleet coordination, and regulatory compliance in future urban AAM systems.