Skip to main content

Autonomous electric vehicle fleets: data-driven analysis of mobility and impact on smart city infrastructure

Primary supervisor

Hao Wang

Autonomous electric vehicle (EV) fleets are emerging as an effective way to solve environmental problems and reduce commute costs in smart cities. Due to the complex spatiotemporal behaviors of passengers and their trips, the unmanaged electric charging demand from EV fleets will significantly impact the existing transportation and electric power infrastructure. Reliable charging networks and charging strategies for EV fleets are the prerequisites to the successful adoption of autonomous EV fleets.

We aim to take the first step to

  • analyze the real-world behaviors of EV fleets, including their mobility patterns and charging demand;
  • design AI-empowered strategies for EV charging and routing to leverage the diversity in the trip patterns and the availability of charging capabilities. The heterogenous EV behaviours and preferences under a dynamical energy-transport environment with uncertainties will be considered.
  • provide insights into charging infrastructure planning.


Student cohort

Double Semester


This project aims to leverage data analytics and machine learning tools to study the real-world mobility patterns and charging demand of EV fleets and to further predict/manage EV travels with smart charging.

  1. You will be provided with realistic trip record data;
  2. This project will reveal the mobility patterns including arrival and departure on a daily basis in a zonal system;
  3. This project will develop methods to estimate the charging demand of EV fleets in each zone/region based on their mobility patterns and manage the EV travels using deep learning.
  4. This project will develop AI solutions for EV charging management.


This thesis project is aligned with the PhD project at

Required knowledge

Programming knowledge (required), Python/R; Data analytics and deep learning (preferred).