Real-World Deployment and Assessment of a Multiagent Reinforcement Learning-Based Variable-Speed-Limit Control System

Yuhang Zhang, Zhiyao Zhang, Junyi Ji, Marcos Quinones-Grueiro, William Barbour, Derek Gloudemans, Gergely Zachar, Clay Weston, Gautam Biswas, Daniel B. Work

Abstract: This article presents the first field deployment of a multiagent reinforcement learning (MARL)-based variable- speed-limit (VSL) control system on Interstate 24 (I-24) near Nashville, TN, USA. We design and demonstrate a full pipeline from training MARL agents in a traffic simulator to a field deployment on a 17-mi segment of I-24 encompassing 67 VSL controllers. The system was launched on 8 March 2024 and has made approximately 35 million decisions on 28 million trips in six months of operation. We apply an invalid action masking mechanism and several safety guards to ensure real-world constraints. The MARL-based implementation operates up to 98% of the time, with the safety guards overriding the MARL decisions for the remaining time. We evaluate the performance of the MARL-based algorithm in comparison to a previously deployed non-RL VSL benchmark algorithm on I-24. The results show that the MARL-based VSL control system achieves a superior performance.

External IDs:doi:10.1109/mits.2025.3592040