Understanding the performance of the human brain to stabilize the body remains an open fundamental research question. In this article, we study the hypothesis of internal model of the Central Nervous System (CNS) by a novel proposed architecture based on a recurrent neural network. The overall objective of the article and the main contribution stems from demonstrating the capability of replicating the balancing mechanisms of the brain by training the proposed bio-inspired network architecture with human balancing data and in the sequel applying the resulting control structure for controlling a single link inverted pendulum. Towards this direction, the body kinetics and kinematics measurements of forty-five subjects during upright stance trails were collected and utilized for training the proposed neural network. The efficacy of the proposed scheme will be proven through multiple simulation results with a single link inverted pendulum, where it will be demonstrated that the brain-inspired control scheme achieves a proper balance. ; ISBN för värdpublikation: 978-3-907144-00-8, 978-1-7281-1314-2


    Access

    Download


    Export, share and cite



    Title :

    Replicating human brain mechanisms towards balancing


    Contributors:

    Publication date :

    2019-01-01



    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Classification :

    DDC:    629




    Mastering/Replicating/Packaging

    British Library Online Contents | 1997


    Toward Self-replicating Robots

    Chirikjian, Gregory S. / Suthakorn, Jackrit | Springer Verlag | 2003


    Media, Mastering, Replicating, & Packaging

    British Library Online Contents | 1998


    Replicating the racing experience

    Automotive engineering | 2015