Multi-agent Reinforcement Learning for Ecological Car-Following Control in Mixed Traffic

Apr 1, 2024·
Qun Wang
,
Fei Ju
,
Huaiyu Wang
,
Yahui Qian
,
Meixin Zhu
,
Weichao Zhuang
,
Liangmo Wang
· 0 min read
Abstract
We propose an ecological car-following control strategy based on multi-agent reinforcement learning (MARL) for connected electrified vehicles in mixed traffic flows. Each CAV agent learns a cooperative policy that balances tracking performance, safety, and energy economy, while adapting to the behavior of human-driven preceding and following vehicles. The proposed method consistently reduces energy consumption over single-agent baselines and demonstrates strong generalization to unseen traffic scenarios.
Type
Publication
IEEE Transactions on Transportation Electrification
publications
Authors
Postdoctoral Fellow
Qun Wang is a Postdoctoral Fellow at the AIMS Lab, Hong Kong Polytechnic University, working with Assistant Professor Hailong Huang. His research focuses on energy-saving optimization and safety-guaranteed control of autonomous electrified vehicles, with emphasis on reinforcement learning for eco-driving, human driving behavior modeling, and multi-agent car-following control. He has published over 10 papers in leading journals including IEEE Transactions on ITS, TTE, TCST and Energy.