Enhancing en-route electric vehicle charging services with AI integration: a collaborative fog-based strategy for optimizing sustainable transportation

Abstract

In the emergence of greener transportation, Electric Vehicles (EVs) play an important role, expected to outnumber conventional vehicles in the near future. However, the installation of Fixed Charging Stations (FCSs) is not keeping up with the increased demand, especially outside urban centers. Such a challenge is prohibiting many users from owning EVs because of range anxiety. This thesis proposes a novel cooperative mechanism where EVs can access charging services such as Vehicle-to-Vehicle (V2V) charging schemes, private smart Home Charging Station (HCS), or Mobile Charging Station (MCS) to complement existing FCS services in certain regions. To this end, the proposed mechanism divides each region into geographically distributed zones managed by cloud-fog nodes for charging service coordination. In each zone, we employ the Hungarian matching algorithm to optimally match EVs with the available charging services. Unlike recent approaches that establish a one-to-one matching between supplier EVs and demanding EVs, our mechanism matches multiple demanding EVs to charging services with a larger capacity to maximize the service offering. Comparing results with existing studies shows that our model outperforms prior approaches across critical factors. Furthermore, our proposed matching algorithm prioritizes EVs requiring charge based on their maximum travel range given their current State of Charge (SoC). To address the challenge of accurately estimating EV driving range, we introduce an ensemblebased Machine Learning (ML) model offering a compelling solution for enhancing the estimation of EV driving range for practical applications.