This paper proposes a linear Gaussian Process learning framework which can be used in the both of semi-parametric and non-parametric models. Inspired by the linear relationship in Euler–Lagrange equation, this approach reduces the dimension of the previous models, which in turn improve the learning efficiency. Besides, the linear relationship results in the better generalization. Simulational results verify the feasibility of the proposed method by predicting the status of a two degrees-of-freedom (DOF) manipulator.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Linear Gaussian Processes for Data-Efficient Robot Dynamics Learning


    Additional title:

    Lect. Notes Electrical Eng.


    Contributors:
    Yan, Liang (editor) / Duan, Haibin (editor) / Yu, Xiang (editor) / Du, Desong (author) / Liu, Changhao (author) / Ni, Chenrui (author) / Qi, Naiming (author) / Liu, Yanfang (author)


    Publication date :

    2021-10-30


    Size :

    11 pages





    Type of media :

    Article/Chapter (Book)


    Type of material :

    Electronic Resource


    Language :

    English




    Linear Gaussian Processes for Data-Efficient Robot Dynamics Learning

    Du, Desong / Liu, Changhao / Ni, Chenrui et al. | TIBKAT | 2022


    Gaussian Processes for Data-Efficient Learning in Robotics and Control

    Deisenroth, MP / Fox, D / Rasmussen, CE | BASE | 2015

    Free access

    Learning Target Dynamics While Tracking Using Gaussian Processes

    Veiback, Clas / Olofsson, Jonatan / Lauknes, Tom Rune et al. | IEEE | 2020



    Sample-efficient robot motion learning using Gaussian process latent variable models

    Delgado-Guerrero, Juan Antonio / Colomé, Adrià / Torras, Carme | BASE | 2020

    Free access