Real-Time Queue Length Estimation with Connected Vehicles Data at Signalized Intersections


The queue length is one of the most useful indicators of congestion on signalized roadways, which plays an essential role in the applications of Intelligent Transport Systems (ITS), such as Advanced Traveller Information Systems and Urban Traffic Control (UTC). Specifically, accurate prediction of queue length is crucial for optimization of traffic signal and vehicle trajectory. In this paper, we investigate the application of shockwave theory in queue length estimation, apply a Maximum Weighted Likelihood Estimator (MWLE) to assign different weights to different samples to achieve more accurate and robust queue length estimation, and discuss potential criteria for use in determining the optimum weights. In closing, further analysis to be explored in the final paper is discussed.

Junyi Li
PhD Student
Simon Hu
Assistant Professor