Experimental comparison of parameter estimation methods in adaptive robot control

Harry Berghuis, Harry Berghuis, Herman Roebbers, Herman Roebbers, Henk Nijmeijer

    Research output: Contribution to journalArticleAcademicpeer-review

    26 Citations (Scopus)
    103 Downloads (Pure)

    Abstract

    In the literature on adaptive robot control a large variety of parameter estimation methods have been proposed, ranging from tracking-error-driven gradient methods to combined tracking- and prediction-error-driven least-squares type adaptation methods. This paper presents experimental data from a comparative study between these adaptation methods, performed on a two-degrees-of-freedom robot manipulator. Our results show that the prediction error concept is sensitive to unavoidable model uncertainties. We also demonstrate empirically the fast convergence properties of least-squares adaptation relative to gradient approaches. However, in view of the noise sensitivity of the least-squares method, the marginal performance benefits, and the computational burden, we (cautiously) conclude that the tracking-error driven gradient method is preferred for parameter adaptation in robotic applications.
    Original languageEnglish
    Pages (from-to)1275-1285
    Number of pages10
    JournalAutomatica
    Volume31
    Issue number9
    DOIs
    Publication statusPublished - 1995

    Fingerprint

    Intelligent robots
    Parameter estimation
    Gradient methods
    Manipulators
    Robotics
    Robots

    Keywords

    • METIS-111802
    • IR-14816

    Cite this

    Berghuis, H., Berghuis, H., Roebbers, H., Roebbers, H., & Nijmeijer, H. (1995). Experimental comparison of parameter estimation methods in adaptive robot control. Automatica, 31(9), 1275-1285. https://doi.org/10.1016/0005-1098(95)00046-Y
    Berghuis, Harry ; Berghuis, Harry ; Roebbers, Herman ; Roebbers, Herman ; Nijmeijer, Henk. / Experimental comparison of parameter estimation methods in adaptive robot control. In: Automatica. 1995 ; Vol. 31, No. 9. pp. 1275-1285.
    @article{10a34977c2f64839946ed831731dd6dd,
    title = "Experimental comparison of parameter estimation methods in adaptive robot control",
    abstract = "In the literature on adaptive robot control a large variety of parameter estimation methods have been proposed, ranging from tracking-error-driven gradient methods to combined tracking- and prediction-error-driven least-squares type adaptation methods. This paper presents experimental data from a comparative study between these adaptation methods, performed on a two-degrees-of-freedom robot manipulator. Our results show that the prediction error concept is sensitive to unavoidable model uncertainties. We also demonstrate empirically the fast convergence properties of least-squares adaptation relative to gradient approaches. However, in view of the noise sensitivity of the least-squares method, the marginal performance benefits, and the computational burden, we (cautiously) conclude that the tracking-error driven gradient method is preferred for parameter adaptation in robotic applications.",
    keywords = "METIS-111802, IR-14816",
    author = "Harry Berghuis and Harry Berghuis and Herman Roebbers and Herman Roebbers and Henk Nijmeijer",
    year = "1995",
    doi = "10.1016/0005-1098(95)00046-Y",
    language = "English",
    volume = "31",
    pages = "1275--1285",
    journal = "Automatica",
    issn = "0005-1098",
    publisher = "Elsevier",
    number = "9",

    }

    Berghuis, H, Berghuis, H, Roebbers, H, Roebbers, H & Nijmeijer, H 1995, 'Experimental comparison of parameter estimation methods in adaptive robot control', Automatica, vol. 31, no. 9, pp. 1275-1285. https://doi.org/10.1016/0005-1098(95)00046-Y

    Experimental comparison of parameter estimation methods in adaptive robot control. / Berghuis, Harry; Berghuis, Harry; Roebbers, Herman; Roebbers, Herman; Nijmeijer, Henk.

    In: Automatica, Vol. 31, No. 9, 1995, p. 1275-1285.

    Research output: Contribution to journalArticleAcademicpeer-review

    TY - JOUR

    T1 - Experimental comparison of parameter estimation methods in adaptive robot control

    AU - Berghuis, Harry

    AU - Berghuis, Harry

    AU - Roebbers, Herman

    AU - Roebbers, Herman

    AU - Nijmeijer, Henk

    PY - 1995

    Y1 - 1995

    N2 - In the literature on adaptive robot control a large variety of parameter estimation methods have been proposed, ranging from tracking-error-driven gradient methods to combined tracking- and prediction-error-driven least-squares type adaptation methods. This paper presents experimental data from a comparative study between these adaptation methods, performed on a two-degrees-of-freedom robot manipulator. Our results show that the prediction error concept is sensitive to unavoidable model uncertainties. We also demonstrate empirically the fast convergence properties of least-squares adaptation relative to gradient approaches. However, in view of the noise sensitivity of the least-squares method, the marginal performance benefits, and the computational burden, we (cautiously) conclude that the tracking-error driven gradient method is preferred for parameter adaptation in robotic applications.

    AB - In the literature on adaptive robot control a large variety of parameter estimation methods have been proposed, ranging from tracking-error-driven gradient methods to combined tracking- and prediction-error-driven least-squares type adaptation methods. This paper presents experimental data from a comparative study between these adaptation methods, performed on a two-degrees-of-freedom robot manipulator. Our results show that the prediction error concept is sensitive to unavoidable model uncertainties. We also demonstrate empirically the fast convergence properties of least-squares adaptation relative to gradient approaches. However, in view of the noise sensitivity of the least-squares method, the marginal performance benefits, and the computational burden, we (cautiously) conclude that the tracking-error driven gradient method is preferred for parameter adaptation in robotic applications.

    KW - METIS-111802

    KW - IR-14816

    U2 - 10.1016/0005-1098(95)00046-Y

    DO - 10.1016/0005-1098(95)00046-Y

    M3 - Article

    VL - 31

    SP - 1275

    EP - 1285

    JO - Automatica

    JF - Automatica

    SN - 0005-1098

    IS - 9

    ER -