A Simulation Framework for Traffic Signal Control under Connected Vehicle Data Environment

Compared with traditional fixed traffic detection, intelligent connected vehicle (CV) brings a new spatial- temporal data environment for urban traffic signal control. However how to evaluate traffic control approaches based on connected vehicle data, especially in different penetration rates environment has been a great difficulty. This paper proposes a simulation framework that incorporates real-world traffic signal controller, system and traffic simulation software, VISSIM. Different penetration rates of connected vehicles could be modeled in VISSIM. We also development a data communication unit (DCU) to exchange real-time information between virtual road networks and real-world traffic signal controller or system via a protocol that we designed. With this simulation framework, researchers could evaluate new traffic signal control strategies based on connected vehicle data, or even multi-source data fusion. Finally, an exemplary traffic control optimization scenario based on CV trajectory data is given to demonstrate the practicability of the simulation framework.