Abstract
The upcoming Connected Vehicles (CV) technology shows great promise in effectively managing traffic congestion and enhancing mobility for users along transportation corridors. Data analysis powered by sensors in Connected Vehicles allows us to implement optimized traffic management strategies optimizing the efficiency of transportation infrastructure resources. In this study, the research team introduces a novel Integrated Corridor Management (ICM) methodology, which integrates underutilized Park-And-Ride (PAR) facilities into the global optimization strategy. To achieve this, the team uses vehicle-to-infrastructure (V2I) communication protocols, namely basic safety messages (BSM) and traveler information messages (TIM) to help gather downstream traffic information and share park and ride advisories with upstream traffic, respectively. Next, the team develops a model that assesses potential delays experienced by vehicles in the corridor. Based on this model, the research team employs a novel centralized deep reinforcement learning (DRL) solution to control the timing and content of these messages. The ultimate goal is to maximize throughput, minimize carbon emissions, and reduce travel time effectively. To evaluate the Integrated Corridor Management strategy, the paper includes simulations on a realistic model of Interstate 5 using the Veins simulation software. The deep reinforcement learning agent converges to a strategy that marginally improves throughput, travel speed, and freeway travel time, at the cost of a slightly higher carbon footprint.
