Abstract for presentation (Poster or Podium)
Intelligent Transportation Systems
Wan Li, PhD (she/her/hers)
Research Staff
Oak Ridge National Laboratory
Knoxville, TN, United States
Wan Li, PhD (she/her/hers)
Research Staff
Oak Ridge National Laboratory
Knoxville, TN, United States
Chieh (Ross) Wang, PhD, A.M.ASCE (he/him/his)
R&D Staff
Oak Ridge National Laboratory
Oak Ridge, TN, United States
Jinghui Yuan, PhD
R&D Staff
Oak Ridge National Laboratory, United States
Hyeonsup Lim, n/a
R&D Associate
Oak Ridge National Laboratory
Oak Ridge, Tennessee, United States
Tim LaClair, n/a
Senior Research Scientist
National Renewable Energy Laboratory, United States
Qichao Wang, PhD (he/him/his)
Researcher III-Computational Science
National Renewable Energy Lab
Golden, CO, United States
Andy Berres, PhD (they/them/theirs)
Research Engineer
National Renewable Energy Laboratory
Golden, CO, United States
hong Wang, PhD
Senior Distinguished R&D Staff
Oak Ridge National Laboratory, United States
Wan Li, PhD (she/her/hers)
Research Staff
Oak Ridge National Laboratory
Knoxville, TN, United States
This paper presents a comprehensive field test of a predictive Bilinear signal control algorithm deployed at three intersections along Shallowford Road corridor in Chattanooga, Tennessee. The primary objective of this research is to assess the algorithm's effectiveness in optimizing signal timings for current and upcoming traffic cycles by maximizing vehicle occupancy on both major and minor streets within the corridor. It is important to note that the algorithm's approach does not involve modifying the existing signal configuration, which still utilizes a dual-ring control system as currently implemented in actuated signal control. Instead, the focus is on optimizing phase splits through predictive bilinear control. Real-world traffic and Signal Phase and Timing (SPaT) information are collected through GRIDSMART sensors and NTCIP servers connected to the signal controllers. The field test was carried out during the morning peak hours from August 8 to August 10, 2023. Performance comparisons were made between the predictive control system and a baseline actuated traffic signal control. The evaluation metrics considered include the average speed within the corridor, throughput, occupancy levels for each approach, and the rates of arrival-on-green light. The findings of this study provide insights into the effectiveness and potential benefits of predictive signal control algorithms in real-world traffic management scenarios, with implications for future smart transportation systems.