Integrated Optimization of Vehicle Speed Control and Traffic Signal Timing: System Development and Testing

Abstract

The research develops an integrated optimal control system to improve the transportation system efficiency and fuel economy on arterial roads by simultaneously optimizing vehicle speeds and traffic signal timings. The proposed approach entails designing two integrated control systems for connected automated vehicles (CAVs) and connected vehicles (CVs), which will be tested in a microscopic traffic simulation software and a driving simulator, respectively. Given that the existing methods are generally complicated, involving high computational cost, the team will start by developing a simple dual-optimization approach and use heuristic algorithms to locate an approximate optimum solution ensuring expedited computations. Meanwhile, the proposed approach will be developed to ensure it can be easily extended from internal combustion vehicles (ICEVs) to other vehicle types such as hybrid electric vehicles (HEVs) or battery electric vehicles (BEVs). Thereafter, the CAV and traffic signal control system will be implemented in a microscopic traffic simulation software so that the system is evaluated in mixed traffic (CAVs and non-CAVs). A simulated traffic network composed of multiple signalized intersections will be used to quantify the system-wide impacts of the proposed CAV/traffic signal control system on traffic mobility, energy consumption and emission levels for various traffic demand and market penetration levels. Lastly, the team will consider the impacts of human errors and perception reaction times (PRTs) when implementing the CV control system in a driving simulator at MSU. The simulator test will be conducted by participants to compare the proposed dual-optimization CV control system with two other scenarios -- adaptive traffic signal control and an eco-driving system previously developed to optimize vehicle trajectories. It is anticipated that the proposed systems will improve the mobility of arterial traffic by reducing delays, energy consumption and vehicle emissions, which are typically higher in low income areas.

Universities and Sponsoring Organizations Involved

Virginia Tech

Morgan State University

Principal Investigators

Hao Chen (VT) Email: haochen@vt.edu
Hesham A Rakha (VT) Email: hrakha@vt.edu
Mansoureh Jeihani (MSU) Email: mansoureh.jeihani@morgan.edu

Funding Sources and Amounts

USDOT: $120,000 (Federal), Virginia Tech: $40,000 (Match), Morgan State University: $20,385 (Match)

Start Date

January 1, 2021

Expected Completion Date

December 31, 2021

Expected Research Outcomes

The proposed research effort will be the first study to develop an integrated optimization of vehicle speed control and signal timing and test in microscopic simulation software and a driving simulator. Considering the approaches developed in previous studies are fairly complicated with high computational cost, the proposed dual optimization algorithm is aimed to be easily implemented into large simulated traffic network and real time implementations. Different from existing studies only designed for ICEVs, the developed algorithm ensures that it can be easily expanded from ICEVs to other vehicle types including HEVs and BEVs.

Expected Equity Impacts and Benefits of Implementation

The proposed CAV and CV control systems will be the first dual optimization system designed for both CAVs and CVs. Considering that currently most vehicles on the road do not have automated control, the proposed CV control system ensures that non-automated control vehicles can also be optimized to drive on arterial roads with the consideration of human errors and perception reaction delay. In addition, the tests in microscopic simulation software and driving simulator will be beneficial for government stakeholders and industry companies to estimate the benefits of implementing the proposed system.

Subject Areas

Connected and Automated Vehicle, Eco-driving, Traffic Signal Control, Integrated Optimization, Microscopic Traffic Simulation, Driving Simulator