Semi-task-dependent and uncertainty-driven world model maintenance

Onderzoeksoutput: Bijdrage aan tijdschriftTijdschriftartikelAcademicpeer review

2 Citaties (Scopus)

Uittreksel

Nearly every task a domestic robot could potentially solve requires a description of the robot’s environment which we call a world model. One problem underexposed in the literature is the maintenance of world models. Rather than on creating a world model, this work focuses on finding a strategy that determines when to update which object in the world model. The decision whether or not to update an object is based on the expected information gain obtained by the update, the action cost of the update and the task the robot performs. The proposed strategy is validated during both simulations and real world experiments. The extended series of simulations is performed to show both the performance gain with respect to a benchmark strategy and the effect of the various parameters. The experiments show the proposed approach on different set-ups and in different environments.

TaalEngels
Pagina's1-15
TijdschriftAutonomous Robots
Volume38
Nummer van het tijdschrift1
DOI's
StatusGepubliceerd - 2014

Vingerafdruk

Robots
Experiments
Uncertainty
Costs

Trefwoorden

    Citeer dit

    @article{ff90ebf80f4c49d2b79dd67736c7b243,
    title = "Semi-task-dependent and uncertainty-driven world model maintenance",
    abstract = "Nearly every task a domestic robot could potentially solve requires a description of the robot’s environment which we call a world model. One problem underexposed in the literature is the maintenance of world models. Rather than on creating a world model, this work focuses on finding a strategy that determines when to update which object in the world model. The decision whether or not to update an object is based on the expected information gain obtained by the update, the action cost of the update and the task the robot performs. The proposed strategy is validated during both simulations and real world experiments. The extended series of simulations is performed to show both the performance gain with respect to a benchmark strategy and the effect of the various parameters. The experiments show the proposed approach on different set-ups and in different environments.",
    keywords = "Information gain, Task dependency, World model verification",
    author = "J. Elfring and {van de Molengraft}, M.J.G. and M. Steinbuch",
    year = "2014",
    doi = "10.1007/s10514-014-9393-0",
    language = "English",
    volume = "38",
    pages = "1--15",
    journal = "Autonomous Robots",
    issn = "0929-5593",
    publisher = "Springer",
    number = "1",

    }

    Semi-task-dependent and uncertainty-driven world model maintenance. / Elfring, J.; van de Molengraft, M.J.G.; Steinbuch, M.

    In: Autonomous Robots, Vol. 38, Nr. 1, 2014, blz. 1-15.

    Onderzoeksoutput: Bijdrage aan tijdschriftTijdschriftartikelAcademicpeer review

    TY - JOUR

    T1 - Semi-task-dependent and uncertainty-driven world model maintenance

    AU - Elfring,J.

    AU - van de Molengraft,M.J.G.

    AU - Steinbuch,M.

    PY - 2014

    Y1 - 2014

    N2 - Nearly every task a domestic robot could potentially solve requires a description of the robot’s environment which we call a world model. One problem underexposed in the literature is the maintenance of world models. Rather than on creating a world model, this work focuses on finding a strategy that determines when to update which object in the world model. The decision whether or not to update an object is based on the expected information gain obtained by the update, the action cost of the update and the task the robot performs. The proposed strategy is validated during both simulations and real world experiments. The extended series of simulations is performed to show both the performance gain with respect to a benchmark strategy and the effect of the various parameters. The experiments show the proposed approach on different set-ups and in different environments.

    AB - Nearly every task a domestic robot could potentially solve requires a description of the robot’s environment which we call a world model. One problem underexposed in the literature is the maintenance of world models. Rather than on creating a world model, this work focuses on finding a strategy that determines when to update which object in the world model. The decision whether or not to update an object is based on the expected information gain obtained by the update, the action cost of the update and the task the robot performs. The proposed strategy is validated during both simulations and real world experiments. The extended series of simulations is performed to show both the performance gain with respect to a benchmark strategy and the effect of the various parameters. The experiments show the proposed approach on different set-ups and in different environments.

    KW - Information gain

    KW - Task dependency

    KW - World model verification

    UR - http://www.scopus.com/inward/record.url?scp=84921067563&partnerID=8YFLogxK

    U2 - 10.1007/s10514-014-9393-0

    DO - 10.1007/s10514-014-9393-0

    M3 - Article

    VL - 38

    SP - 1

    EP - 15

    JO - Autonomous Robots

    T2 - Autonomous Robots

    JF - Autonomous Robots

    SN - 0929-5593

    IS - 1

    ER -