Bounding is one of the important gaits in quadrupedal locomotion for negotiating obstacles. The authors proposed an effective approach that can learn robust bounding gaits more efficiently despite its large variation in dynamic body movements. The authors first pretrained the neural network (NN) based on data from a robot operated by conventional model-based controllers, and then further optimised the pretrained NN via deep reinforcement learning (DRL). In particular, the authors designed a reward function considering contact points and phases to enforce the gait symmetry and periodicity, which improved the bounding performance. The NN-based feedback controller was learned in the simulation and directly deployed on the real quadruped robot Jueying Mini successfully. A variety of environments are presented both indoors and outdoors with the authors’ approach. The authors’ approach shows efficient computing and good locomotion results by the Jueying Mini quadrupedal robot bounding over uneven terrain.


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    Efficient learning of robust quadruped bounding using pretrained neural networks


    Beteiligte:
    Wang, Z (Autor:in) / Li, A (Autor:in) / Zheng, Y (Autor:in) / Xie, A (Autor:in) / Li, Z (Autor:in) / Wu, J (Autor:in) / Zhu, Q (Autor:in)

    Erscheinungsdatum :

    25.09.2022


    Anmerkungen:

    IET Cyber-Systems and Robotics (2022) (In press).


    Medientyp :

    Aufsatz (Zeitschrift)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch



    Klassifikation :

    DDC:    629





    Three-Rigid-Body Model Based NMPC for Bounding of a Quadruped with Two Spinal Joints

    Huang, Songrui / Cai, Wenhan / Zhao, Mingguo | TIBKAT | 2024


    Using Pretrained AlexNet Deep Learning Neural Network for Recognition of Underwater Objects

    Szymak, Piotr / Gasiorowski, Marek | BASE | 2020

    Freier Zugriff

    Robust ellipsoidal state bounding algorithm

    Wei, C. / Xianfang, S. | British Library Online Contents | 2006