Integrating AI for Autonomous UAV Traffic Management in Drone Logistic Operations: Challenges, Approaches, and Future Directions

As urban airspaces become increasingly congested, managing autonomous UAV operations presents significant challenges, particularly in ensuring safety and efficiency through effective flight plan approval and conflict resolution mechanisms. This paper addresses these challenges by investigating the role of artificial intelligence (AI) in optimizing UAV operations within U-Space, focusing on strategic deconfliction and autonomous flight plan approvals. The primary aim of this study is to explore how AI techniques - specifically machine learning, optimization algorithms, and explainable AI (XAI) - can be integrated into UAV traffic management systems to support safe, scalable, and fair urban airspace use. The AI4HyDrop project is used to illustrate an AI-based approach in supporting fair and environmentally sustainable airspace use, addressing challenges in strategic deconfliction and airspace allocation. Ultimately, this work seeks to answer the question: "How can AI techniques be integrated to optimize autonomous drone flight plan approvals with mechanisms for strategic deconfliction in high-density airspaces?"By emphasizing the potential of AI to transform urban UAV operations, this study outlines key directions for research and collaboration in developing safer and more efficient drone logistic operations.