The development of 5G has enabled the autonomous vehicles (AVs) to have full control over all functions. The AV acts autonomously and collects travel data based on various smart devices and sensors, with the goal of enabling it to operate under its own power. However, the collected data is affected by several sources that degrade the forecasting accuracy. To manage large amounts of traffic data in different formats, a computational data science approach (CDS) is proposed. The computational data science scheme introduced to detect anomalies in traffic data that negatively affect traffic efficiency. The combination of data science and advanced artificial intelligence techniques, such as deep leaning provides higher degree of data anomalies detection which leads to reduce traffic congestion and vehicular queuing. The main contribution of the CDS approach is summarized in detection of the factors that caused data anomalies early to avoid long-term traffic congestions. Moreover, CDS indicated a promoting results in various road traffic scenarios.


    Access

    Download


    Export, share and cite



    Title :

    Detection of road traffic anomalies based on computational data science


    Additional title:

    Discov Internet Things


    Contributors:

    Published in:

    Publication date :

    2022-10-20




    Type of media :

    Article (Journal)


    Type of material :

    Electronic Resource


    Language :

    English






    A Computational Data Science Based Detection of Road Traffic Anomalies

    Murali Krishna, Chinta Venkata / Gopala Krishna, G / Vellela, Sai Srinivas et al. | IEEE | 2023



    Multidimensional Data Mining of Traffic Anomalies on Large-Scale Road Networks

    Gonzalez, Hector / Han, Jiawei / Ouyang, Yanfeng et al. | Transportation Research Record | 2011