Researchers from the University of Hong Kong, including Aoyu Pang, Maonan Wang, Zifan Sha, Wenwei Yue, Changle Li, Chung Shue Chen, and Man-On Pun, have developed a novel approach to optimize urban air mobility (UAM) services, aiming to integrate air and ground transportation for more efficient door-to-door travel. Their work, titled “Heterogeneous Vertiport Selection Optimization for On-Demand Air Taxi Services: A Deep Reinforcement Learning Approach,” was recently published in the journal Transportation Research Part C: Emerging Technologies.
The team recognized that while UAM, or air taxis, could help alleviate urban congestion by utilizing low-altitude airspace, current research lacks a systematic approach to integrate air and ground transportation for optimal routing. To bridge this gap, they proposed a unified optimization model that considers both air and ground transportation strategies, incorporating real-time traffic conditions and passenger behavior.
Building on this model, the researchers developed the Unified Air-Ground Mobility Coordination (UAGMC) framework. This framework uses deep reinforcement learning (RL) and Vehicle-to-Everything (V2X) communication to optimize vertiport selection and dynamically plan air taxi routes. The UAGMC framework dynamically adjusts to changing conditions, aiming to provide the most efficient travel routes for passengers.
Experimental results showed that the UAGMC framework achieved a 34% reduction in average travel time compared to conventional proportional allocation methods. This improvement highlights the potential of the UAGMC framework to enhance travel efficiency and provide valuable insights into the integration and optimization of multimodal transportation systems. The researchers have made the related code available on GitHub for further exploration and development.
For the energy industry, this research could have practical applications in optimizing energy-efficient routes for air taxis, reducing overall energy consumption, and minimizing emissions. As UAM services become more prevalent, such advancements could contribute to more sustainable and efficient urban transportation networks. The research was published in the journal Transportation Research Part C: Emerging Technologies.
This article is based on research available at arXiv.