One of the challenges facing intersection safety systems is accurate positioning or localisation of a vehicle at an intersection. Global Positioning Systems (GPS) are obvious candidates to address this issue, but in built-up areas GPS measurement accuracy and availability can be poor. In response, TRW Conekt has built on its Video Lane Detection technology and Data Fusion experience to develop a Video- Based Intersection Localisation system. This paper gives a brief description of the position localisation system concept and provides details about the real-time implementation of the system. The results that have been observed are then presented before concluding with a review of the system performance and future enhancements that could be made. The localisation system design is based upon the concept of comparing a low-level feature map of known visual landmarks with a processed image from a video camera mounted in the vehicle. The vehicle's position and motion with respect to the map can be estimated by tracking the measurements using some form of tracking filter. Assuming that the position of the vehicle is approximately known, the system can predict the position of lines on the road surface for the next image captured. The position of each line is then measured by searching for the predicted lines in the imager output. Each predicted line then has a measurement associated with it. Once the lines have been associated, the error between the prediction and measurement can be determined. The error is used to correct the position estimate of the vehicle within the intersection. This new vehicle position estimate is then used to predict the position of the lines in the next image ready for the next iteration of the localisation algorithm. The video based localisation system has been integrated successfully into the main INTERSAFE test vehicle. In this integrated system, the vehicle speed and yaw rate are taken directly from the vehicle CAN bus. The output of the video based localisation system is fused with the output from a laser-scanner based localisation system. This fused output is then used in high-level systems that warn the driver about dangerous situations within the intersection.


    Access

    Access via TIB

    Check availability in my library


    Export, share and cite



    Title :

    Video based intersection localisation


    Additional title:

    Videobasierte Egopositionserkennung in Kreuzungsbereichen


    Contributors:
    Tucker, Mark (author) / Heenan, Adam (author)


    Publication date :

    2006


    Size :

    13 Seiten, 5 Bilder, 8 Quellen


    Type of media :

    Conference paper


    Type of material :

    Print


    Language :

    English




    Video based intersection localisation

    Tucker,M. / Heenan,A. / TRW Conekt,Solihull,GB | Automotive engineering | 2006


    Video-based intersection area traffic parameter statistical method

    CHEN YANGZHOU / SHI ZEYU / XIN LE et al. | European Patent Office | 2020

    Free access

    Video data based Traffic State Prediction at Intersection

    Shi, Zeyu / Chen, Yangzhou / Ma, PengFei | IEEE | 2020


    Video-based intersection deadlock event detection method and device

    CHEN WEIQIANG / WANG WENWEN / CHEN XIAOMING et al. | European Patent Office | 2022

    Free access

    Video information based method for detecting vehicles at intersection

    CAI YANGUANG / LIU HONGYU / CAI HAO et al. | European Patent Office | 2015

    Free access