This paper studies the problem of real-time traffic estimation and incident detection by posing it as a hybrid state estimation problem. An interactive multiple model ensemble Kalman filter is proposed to solve the sequential estimation problem, and to accommodate the switching dynamics and nonlinearity of the traffic incident model. The effectiveness of the proposed algorithm is evaluated through numerical experiments using a perturbed traffic model as the true model. The supporting source code is available for download at https://github.com/Lab-Work/IMM_EnKF_Traffic_Estimation_Incident_Detection.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Interactive multiple model ensemble Kalman filter for traffic estimation and incident detection


    Contributors:
    Wang, Ren (author) / Work, Daniel B. (author)


    Publication date :

    2014-10-01


    Size :

    1278343 byte




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English