Autonomous 4D Trajectory Planning for Dynamic and Flexible Air Traffic Management

With an ever increasing number of unmanned aerial vehicles (UAVs) in flight, there is a pressing need for scalable and dynamic air traffic management solutions that ensure efficient use of the airspace while maintaining safety and avoiding mid-air collisions. To address this need, a novel framework is developed for computing optimized 4D trajectories for UAVs that ensure dynamic and flexible use of the airspace, while maximizing the available capacity through the minimization of the aggregate traveling times. Specifically, a network manager (NM) is utilized that considers UAV requests (including start/target locations) and addresses inherent mobility uncertainties using a linear-Gaussian system, to compute efficient and safe trajectories. Through the proposed framework, a family of mathematical programming problems is derived to compute control profiles for both distributed and centralized implementations. Extensive simulation results are presented to demonstrate the applicability of the proposed framework to maximize air traffic throughput under probabilistic collision avoidance guarantees.