SOURCE: Towards Solar-Uncertainty-Aware E-Taxi Coordination under Dynamic Passenger Mobility

As more fuel-based vehicles are replaced by electric vehicles (EVs) for providing transportation services in cities, e.g., electric taxi or bus fleets, solar-powered charging stations have been progressively deployed in the cities to provide charging services for EV fleets. The emerging solar-powered charging stations bring opportunities for EV fleets to reduce charging cost and carbon footprint by utilizing cheap and environment-friendly solar energy. The uncertainty and dynamics of solar power, however, raise challenges in making full use of solar power on time for EV fleets, e.g., e-taxis, while at the same time serving passengers efficiently. In this work, we design SOURCE, a solar-uncertainty-aware coordination algorithm for e-taxis under spatial-temporal dynamics of uncertain solar energy and passenger mobility. We evaluate our solution with a comprehensive dataset for an existing e-taxi system and charging infrastructures including 726 e-taxis, 7,228 regular fuel taxis, 37 working charging stations, and 62,100 collected taxi trips per day. Our data-driven evaluation shows that our solution significantly improves the usage rate of solar power by 17.6% with similar passenger waiting time compared to the solution that co-optimizes service quality and usage of solar power without considering the solar power uncertainty.