This paper proposes a method of robot autonomous navigation along routes by landmarks. A robot continuously scans the side-view and yields an intermediate representation called a panoramic representation. The objects in the scene are segmented based upon range and color information of the panoramic representation and then, are arranged into a structure map which provides robust qualitative information on the environment structure. The landmarks are selected as objects in the structure map with distinctive features along the route. When the robot traverses along the route again with a changed speed, objects in the new scene are segmented and arranged in a new structure map. Since the relative spatial arrangement is invariant with the camera speed, the landmarks are matched with the objects in the new structure map by dynamic programming method. Further, the correctness of matching is verified using the disparity and color cues.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Route description by landmarks


    Contributors:
    Shigang Li (author) / Ochi, A. (author) / Tsuji, S. (author)


    Publication date :

    1995-01-01


    Size :

    602469 byte




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Route Description by Landmarks

    Li, S. / Ochi, A. / Tsuji, S. et al. | British Library Conference Proceedings | 1995


    3D human face description: landmarks measures and geometrical features

    Vezzetti, E. / Marcolin, F. | British Library Online Contents | 2012


    The use of landmarks for enhanced route guidance: The REGIONAL project

    Burnett, G. / ITS Congress Association | British Library Conference Proceedings | 2000


    LOCALIZATION USING DYNAMIC LANDMARKS

    LIN SÖRSTEDT JOAKIM / SCHINDLER ANDREAS / GUSTAFSSON TONY | European Patent Office | 2024

    Free access

    Route and Traffic Description

    Sciarretta, Antonio / Vahidi, Ardalan | Springer Verlag | 2019