We consider the problem of estimating queue-lengths at an intersection from a pair of advance and stop bar detectors that count vehicles, when these measurements are noisy and biased. The key assumption is that we know weather the queue is empty or not. We propose a real-time queue estimation algorithm based on stochastic gradient descent. The algorithm provably learns the detector bias, and efficiently estimates the queue-length with theoretical guarantee. The algorithm is tested in a simulation and in a case study using traffic data from an intersection in Beaufort, North Carolina.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Queue-length estimation using real-time traffic data


    Contributors:


    Publication date :

    2016-11-01


    Size :

    520804 byte





    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Real-time queue length estimation using event-based advance detector data

    An, Chengchuan / Wu, Yao-Jan / Xia, Jingxin et al. | Taylor & Francis Verlag | 2018


    Real-Time Queue Length Estimation for Signalized Intersections Using Vehicle Trajectory Data

    Li, Fuliang / Tang, Keshuang / Yao, Jiarong et al. | Transportation Research Record | 2017


    Lane-based real-time queue length estimation using license plate recognition data

    Zhan, Xianyuan / Li, Ruimin / Ukkusuri, Satish V. | Elsevier | 2015