AUTHOR=Xu Changxian , Sun Zhongbo , Wang Chen , Wu Xiujun , Li Binglin , Zhao Liming 

TITLE=An advanced bionic knee joint mechanism with neural network controller

JOURNAL=Frontiers in Neurorobotics

VOLUME=Volume 17 - 2023

YEAR=2023

URL=https://www.frontiersin.org/journals/neurorobotics/articles/10.3389/fnbot.2023.1178006

DOI=10.3389/fnbot.2023.1178006

ISSN=1662-5218

ABSTRACT=In this paper, a tensegrity based knee mechanism is investigated for the purpose of developing a high efficiency rehabilitation knee exoskeleton. Besides, the kinematics and dynamics models of the knee mechanism are explored for further controller design. In addition, for the purpose of studying the bionic knee joint performances, based on the limit function of knee patella, the limit functionality of the bionic knee joint is developed for enhancing the bionic property. Furthermore, for eliminating the noise item and other disturbances that certainly generated in the rehabilitation process, a noise-tolerant zeroing neural network (NTZNN) algorithm is utilized to establish the controller. Which means the controller has the performance of anti-noise, hence, it is quite unique with other bionic knee mechanism controller. Eventually, the anti-noise performance and calculating precision of the NTZNN controller are verified through several simulations and contrast results.