Modeling Methodology for the Research
With the increasing demand for real-time control logic, infrastructure-enabled vehicle detector data is being considered for state-of-the-art traffic signal control strategies to monitor dynamic arrivals of vehicles approaching an adaptive signalized intersection. Conventional detection methods, typically point detection, cannot directly measure vehicle speed and location, which remains as an obstacle in the design of a robust adaptive traffic control system. The emerging Connected Vehicle (CV) technology provides an opportunity to formulate a mobile data platform that allows for data transfer among multiple vehicles as well as with the roadside infrastructure. The CV’s capability of transmitting the mobile trajectory data to control systems could help enhance the control efficiency while reducing dependencies on conventional infrastructure-based vehicle detectors. The challenge to achieve this envisioned cooperation lies in enabling the connectivity and interoperability of the CV mobile data and the signal control infrastructure.
To address this challenge, the cooperative adaptive signal timing optimization (CASTO) algorithm has been developed to optimize the signal timing at both isolated intersections and intersections along a corridor. The algorithm uses models to process and transmit the CVs’ trajectory datasets (including second-by-second vehicle locations and instantaneous speeds) between CVs on-board units, roadside equipment, and signal controllers to optimize the signal timing at the signalized intersection. A simulation-based platform was created in VISSIM to test the methodology and the CASTO algorithm. A corridor consisting of four signalized intersections was evaluated at four different CV penetration rates under three different traffic conditions, i.e., light traffic, mild traffic, and heavy traffic conditions. The simulation test results show that average vehicle delay and queue length with the CASTO algorithm was reduced by 46.04% and 56.15%, respectively, with 50% penetration of CVs, compared to existing signal control algorithms with 100% conventional vehicles.
This research/study is a foundation for the development of a cooperative traffic control ecosystem in the future. The real-time detection of CV trajectories, combined with data from traditional detection systems, provides increased potential to foster improvements to intelligent transportation infrastructures to improve mobility performance from the perspectives of travel time reduction, enhanced driving comfort, and reduced fuel consumption and emission.
Learning Outcomes:
a) Identify and define the benefits of the Connected Vehicle Ecosystem.
b) Understand how the real-time detection of CVs/CAVs in combination with the CASTO algorithm can provide unprecedented opportunities to improve traffic performance from the aspects of travel time reduction, enhancing driving comfort, and reducing fuel consumption and emissions.
Real-Time Optimization of Traffic Control Using Connected Vehicle Data
