Signalized arterials cause stop-and-go traffic, leading to collisions, delays, energy waste, and discomfort. Connected automated vehicles (CAVs) offer a solution through cooperative eco-driving, but mixed traffic complicates implementation. In this presentation, we introduce rule-based, optimization-based, and reinforcement learning (RL)-based eco-driving systems for CAVs on signalized arterials, balancing safety, efficiency, energy saving, and comfort. These methods include eco-GLOSA, eco-ACC (DDPG, PPO, SAC), and eco-CACC (MADDPG, MASAC). Evaluations use naturalistic pNEUMA trajectory data, benchmarking against human-driven and car-following (CF) models in diverse scenarios. The study comprehensively assesses method effectiveness, platoon stability, and key influencing factors in real-world-inspired conditions, thus delivering actionable and transferable solutions for implementing cooperative automated systems in mixed traffic with current and near-future infrastructure.

Zhiwei Yang is a Postdoctoral Researcher at The Chinese University of Hong Kong, Shenzhen. She received her PhD degree from The University of Queensland in 2025. Her research focuses on intelligent transport systems, urban mobility, reinforcement learning applications for motion control, and High-speed Rail. She has authored multiple papers in Transportation Research Part A, C, and D, Transport Policy, the Transportation Research Board (TRB) Annual Meetings, the IEEE Intelligent Vehicle Symposium (IV), the Australasian Transport Research Forum (ATRF), and the International Conference of Hong Kong Society for Transportation Studies (HKSTS). She also serves as a reviewer for journals such as IEEE Transactions on Intelligent Transportation Systems (TITS), Transportation Research Part C and TRB.