The hidden energy demand of automated mobility: onboard and offboard systems in urban fleet simulations

Abstract

This study examines the uncertain energy impacts of automated on-demand transportation, considering not only vehicle propulsion but also the often-overlooked energy use of onboard and offboard systems required for automation, as well as behavioral changes in mobility patterns that introduce additional complexity. To address these challenges, the research combines agent-based demand modeling with a dynamic fleet controller based on shareability graph principles and cluster analysis, enabling large-scale urban simulations that capture both user behavior and system-level operations. The framework provides a detailed representation of interactions between travel demand, fleet management, and service configuration under different operating conditions. The study further incorporates insights from automated vehicle manufacturers compiled in a European Commission report to derive per-kilometer energy consumption coefficients for both onboard sensing and computing systems and offboard data processing infrastructure. Applied to the Tel Aviv metropolitan area, the simulation evaluates multiple deployment scenarios, including RoboTaxis and automated shuttles, under varying demand levels and service strategies. Results indicate that overall energy demand depends not only on fleet size and travel demand, but also on technological configurations and user preferences. The findings highlight significant hidden energy costs associated with automation, which can offset expected efficiency gains from electrification, and show that prioritizing user convenience over system efficiency limits ride-sharing potential and reduces opportunities for energy savings.