Hybrid AI-Based Demand-Capacity Balancing for UAS Traffic Management and Urban Air Mobility

With the gradual diffusion of commercial Unmanned Aircraft Systems (UAS) operations, UAS transportation and Urban Air Mobility (UAM) services are expected to thrive at low altitudes in cities. The operation of multiple manned and unmanned aircraft may cause airspace capacity overload in dense metropolitan regions in the future. Such congestion and imbalance will reduce the efficiency and safety of operations. Therefore, UAS Traffic Management (UTM) systems will crucially need to provide Demand Capacity Balancing (DCB) services for low-altitude airspace to reduce the criticality of human operators' intervention. This paper proposes a UTM system framework based on a hybrid Artificial Intelligence (AI) algorithm, which supports a resilient and flexible DCB process and solution framework, hence meeting the stringent operational requirements of urban low-altitude airspace. The hybrid AI algorithm includes a training data generation component which trains and optimizes the decision-making model, improving the decision-making performance of the system. A preliminary verification case study is presented, highlighting the capability of the system to generate multiple feasible solutions to airspace congestion problems.