BRIDGING GROUND AND AIR MOBILITY: INTEGRATED VEHICLE ASSIGNMENT AND PATH PLANNING

This dissertation addresses the growing need for efficient and sustainable urban mobility by proposing advanced optimization models for multimodal transportation systems that in- tegrate air taxis and ground vehicles. The study begins by analyzing recent path planning algorithms for unmanned aerial vehicles (UAVs) to understand their applicability in urban air mobility (UAM) scenarios. Building on this analysis, a novel Mixed Integer Nonlin- ear Programming (MINLP) model is developed to jointly optimize vehicle assignment and path planning across both ground and aerial networks. The model incorporates real-time user demand, fleet capacity, energy consumption, and quality-of-service constraints using a rolling-horizon batch framework.

To further improve computational performance and enable scalability, a Mixed Integer Linear Programming (MILP) model is proposed. Comparative experiments are conducted to evaluate both models across a range of metrics including service rate, vehicle usage rate, CO emissions, CPU time, and waiting time cost under different system scales and modal compo- sitions. Results show that the integration of air taxis into ground transportation networks significantly improves operational efficiency and sustainability, particularly in medium- and large-scale scenarios. The MILP model demonstrates superior computational efficiency while maintaining high-quality solutions, reducing average waiting time costs by up to 23.3% and CO emissions by over 50%.

This work represents one of the first optimization frameworks to holistically address vehicle assignment and path planning in an integrated air-ground mobility-on-demand (MoD) system. The findings lay the foundation for future urban transportation strategies that combine aerial and terrestrial modes to balance performance, scalability, and environmental impact.