Abstract
Traffic signal optimization is a significant means for smoothing urban traffic flow. However, the operation of traffic signals is currently seriously constrained by the data available from traditional point detectors. In recent years, an emerging technology, connected vehicle (CV), which can percept the overall traffic environment in real time, has drawn researchers’ attention. With the new data source, traffic controllers should be able to make smarter decisions. A lot of work has been done to develop a new traffic signal control pattern under connected-vehicle environment. This paper provides a comprehensive review of these studies, aiming at sketching out the state of the arts in this research field. Several basic control problems, communication, control input, and objectives, are briefly introduced. The commonly used optimization models for this problem are summarized into three types: rule-based models, mathematical programming-based models, and artificial intelligence-based models. Then some major technical issues are discussed in detail. Finally, we raise the limitation of the existing studies and give our perspectives of the future research directions.
Highlights
To solve the aforementioned problems, one may consider expanding the road facilities to satisfy the increasing travel demand
For the reason that the timing plan is designed based on historical data in an offline manner, this control strategy is known as offline control. e underlying assumption of fixed-time control is that the traffic demand remains unchanged within the entire time period of a timing plan
It has been proved that the introduction of connected vehicle (CV) technology into traffic signal control has potential in improving the road traffic flow efficiency, providing enhanced safety, saving energy consumption, and reducing pollutant emissions. e paper reviews the existing studies about time signal optimization under CV environment to illustrate the current state of the art in this research field
Summary
To solve the aforementioned problems, one may consider expanding the road facilities to satisfy the increasing travel demand. The in-use traffic control strategies can be categorized into three types: (1) fixed-time, (2) actuated, and (3) adaptive control. (2) Actuated control: Actuated control detects the dynamic traffic demand to modify a fixed timing plan by occasionally skipping a phase if no vehicle is present or shortening a phase when vehicles are not being served. Erefore, actuated and adaptive control systems are increasingly adopted in metropolises in order to deal with the downsides of fixed-time control. All the aforementioned actuated and adaptive signal control systems collect real-time traffic data from infrastructurebased sensors, for example, loop detectors, ultrasonic detectors, or video detectors, which can only conduct point detection and estimate traffic states based on very limited information, such as vehicle counts or temporal gap between consecutive vehicles. The existing systems are not able to collect traffic information comprehensively, steadily, and at low cost
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
