Modeling of CAV-based variable speed limit strategy for expressway bottlenecks: An improved cell transmission model

The advancement of Connected and Automated Vehicles (CAVs) technology enables precise control over vehicle trajectories and dynamic adjustment of vehicle states, offering new approaches to traffic management. This study proposes a novel Variable Speed Limit (VSL) control strategy for expressway bottleneck sections, leveraging CAVs as "moving barriers" to implement VSL control in mixed traffic flow (CAV-based VSL), where CAVs coexist with Human-Driven Vehicles (HDVs). The core idea of this strategy is to proactively control CAVs to decelerate, thereby enabling the following HDVs to reduce speed within the designated control sections. To implement and validate this strategy, an Improved Cell Transmission Model (ICTM) is developed through analyzing vehicle microscopic behaviors under the CAV-based VSL control strategy. The ICTM facilitates rapid calculations and effective strategy implementation of CAV-based VSL strategy, establishing a correlation between microscopic vehicle interaction behaviors and the macroscopic model in mixed traffic flow. Simulation experiments are conducted to assess the effectiveness of the CAV-based VSL control strategy in comparison to the No-VSL, MPC-based, RL-based, and HDV-based VSL strategies under varying CAV market penetration rates. The results demonstrate that the CAV-based VSL outperforms the other strategies, reducing total queue time by 42.02 % and the probability of congestion by 41.42 % compared to the No-VSL. Furthermore, the CAV-based VSL significantly reduces the speed variance, whereas the other VSL show no significant mitigation in this metric. Furthermore, sensitivity and quantitative analyses of model parameter selection provide additional validation of the effectiveness and robustness of the proposed CAV-based VSL control strategy.