A Federated Simulation Platform for the Evaluation of Connected Vehicle Applications

Abstract

Connected Vehicle (CV) technology is a promising advancement that has the potential to improve traffic safety and traffic efficiency, and support other Intelligent Transportation Systems (ITS) applications through facilitating vehicle-to-vehicle and vehicle-to-infrastructure communication. Since conducting a CV field test is costly, hazardous and uncontrollable, simulation becomes a preferable method for CV evaluation. Over the past decade, there have been several approaches proposed to implement a CV simulation platform; however, each approach has specific limitations. Most existing simulation platforms are not capable of supporting accurate, complex and large-scale applications. In response, this thesis proposes a new CV simulation platform, which is a federation of a commercial traffic simulator, VISSIM, and an open-source wireless network simulator, OMNeT++. This new platform supports large-scale simulations and comprehensive driving behaviors with high accuracy. Several traffic scenarios were evaluated under the dedicated short-range communication (DSRC) protocol to explore network latency issues. The findings reveal that network latency may become a significant issue when many vehicles attempts to communicate simultaneously. The research herein also evaluated advisory driving speed (ADS) in a CV environment. The results show the potential of CV technologies to solve both recurrent and non-recurrent bottleneck problems.