Humanoid Air Traffic Coordinators for Vertiport Operations in High-Density Urban Skies

The problem of traffic congestion in megacities is increasing annually, helicopters are expensive, noisy, and polluting, cities lack space for new infrastructure, emergency medical transport is slow, and air pollution requires a zero-carbon alternative. Urban Air Mobility (UAM) using electric Vertical Take-Off and Landing (eVTOL) air taxis provides a clean and fast answer, but vertiport operations of hundreds of air vehicles lack a predictive AI system for coordinated air vehicles and social humanoid robots for passenger, baggage, and crowd interactions. This project solves the problem by proposing a novel framework for the fusion of scalable multi-agent reinforcement learning for the conflict-free coordination of 100+ eVTOL air vehicles, multi-modal IoT sensor fusion, and social humanoid robots using reinforcement learning inside a realistic digital twin. Literature survey of 50+ sources (2017-2025) reveals a lack of scalability, humanoid robots, and simulation realism. Three testable hypotheses were formulated. Key findings of the project include the novel architecture for 80-120 flights/hour with zero crashes and increased passenger trust. Applications of the project include sustainable and inclusive urban mobility for megacities like Muscat, Singapore, and New York, thus advancing the LGPR Sustainable Global Societies Initiative.