V2X-Enabled Air-Ground Traffic Coordination for Enhancing On-Demand Air-Taxi Mobility

Urban Air Mobility (UAM) offers a promising solution to urban congestion by utilizing low-altitude airspace, effectively alleviating pressure on ground transportation. The integration of air-taxi services with existing ground transport infrastructure enables streamlined, efficient door-to-door travel. However, current research on air-taxi systems often overlooks the role of passenger decision-making in selecting the optimal boarding vertiport, a factor that can greatly impact overall system efficiency and user satisfaction. To address this problem, we propose a Unified Air-Ground Mobility Coordination (UAGMC) framework. This framework, powered by deep reinforcement learning (RL) and Vehicle-to-Everything (V2X) communication, optimizes vertiport selection and dynamically plans air-taxi routes based on real-time air and ground traffic conditions. Experimental results show that our approach reduces average travel time by 32 % compared to traditional allocation methods using proportional distribution. This framework advances overall travel efficiency and provides novel insights for integrating multimodal transportation systems.