The dangers of and the time needed when clearing an area from unexploded ordnance can be reduced by a system consisting of unmanned, autonomous robots. The system will need less time when more than one robot cooperates to search the area. The reinforcement learning algorithm GPOMDP is evaluated for the specific case of finding a decision rule that, given a map and the robot's position on the map, enables the robot to automatically choose between different possible actions. The actions lead to a near optimal path through an area where some parts need to be searched. A neural network is used as a function approximator to store and improve the decision rule, and also to find actions according to it. The problem expanded to include two robots using the same decision rule, distributed in a sense that the robots pick actions according to their own perception of the surroundings and independent of the other robot's action. To achieve cooperation between the robots, they are trained to maximize a shared reward that is equal to the sum of individual rewards that are given according to the consequences of the robots' actions. When using the learnt policy to search the largest of the experiment's areas, two robots that have been trained with a shared reward use 70% of the time that one optimal robot would need, while two agents that have been trained with their individual rewards need 88%.


    Zugriff

    Zugriff über TIB

    Verfügbarkeit in meiner Bibliothek prüfen


    Exportieren, teilen und zitieren



    Titel :

    Learning to Cooperate in a Search Mission via Policy Search


    Beteiligte:
    D. Martin (Autor:in)

    Erscheinungsdatum :

    2002


    Format / Umfang :

    62 pages


    Medientyp :

    Report


    Format :

    Keine Angabe


    Sprache :

    Englisch




    Search for Extraterrestrial Planets: The Darwin Mission

    D Arcio, L. / European Space Agency / Centre Nationale d'Etudes Spatiales | British Library Conference Proceedings | 2006


    Kepler Mission: A Search for Habitable Planets

    Koch, David / Fonda, Mark | NTRS | 2002



    OSIRIS-REx Extended Mission Trajectory Design & Target Search

    Sutter, Brian / Hatten, Noble / Getzandanner, Kenneth M. et al. | AIAA | 2022


    The Darwin mission: Search for extra-solar planets

    Kaltenegger, L. / Fridlund, M. | Elsevier | 2005