This thesis describes an approach for cooperative multi-robot localization based on probabilistic method (Monte Carlo Localization) used in assistant robots which are capable of sensing and communicating one with another. In our approach, each of the robots maintains its own clustering based MCL algorithm, and communicates with each other whenever it detects another robot. We develop a new information exchange mechanism, which makes use of the information extracted from the clustering component, to synchronize the beliefs of detected robots. By avoiding unnecessary information exchange whenever detection occurs through a belief comparison, our approach can solve the delayed integration problem to improve the effectiveness and efficiency of multi-robot localization. This approach has been tested in both real and simulated environments. Compared with single robot localization, the experimental results demonstrate that our approach can notably improve the performance, especially when the environments are highly symmetric.


    Access

    Download


    Export, share and cite



    Title :

    An Improved Clustering based Monte Carlo Localization approach for Cooperative Multi-robot Localization


    Contributors:

    Publication date :

    2011-01-01


    Remarks:

    Electronic Theses and Dissertations


    Type of media :

    Theses


    Type of material :

    Electronic Resource


    Language :

    English



    Classification :

    DDC:    629




    Localization of Indoor Mobile Robot Using Monte Carlo Localization Algorithm (MCL)

    Mahmood, Ali Khaleel / Bicker, Robert | BASE | 2016

    Free access


    An Improved Approach For Multi-Robot Localization

    Su, Huaicheng | BASE | 2008

    Free access

    Dynamic Maps in Monte Carlo Localization

    Milstein, A. / Canadian Society for Computational Studies of Intelligence | British Library Conference Proceedings | 2005