Abstract
This paper describes a control algorithm for the sliding mode of a lower-extremity exoskeleton model with a neural network for elimination of external perturbances and uncertainties. The sliding control of rapid action is used for achievement of rapid convergence in finite time, absence of singularity, and suppression of oscillations. The neural network allows the efficiency of the regulator inside the boundary layer to be improved. The analysis of the asymptotic stability of the closed system is confirmed by the Lyapunov stability criterion guaranteeing the condition of sliding. The efficiency of the proposed control method has been proven using simulation in the MATLAB software, including the construction of a five-joint mathematical model of a lower-extremity exoskeleton.
REFERENCES
Truong, D.D., Belov, M.P., and Phuong, T.H., Development of mathematical model and subordinate control for nonlinear electric drivers of exoskeleton, IV Int. Conf. on Control in Technical Systems (CTS), St. Petersburg, 2021, IEEE, 2021, pp. 131–134. https://doi.org/10.1109/CTS53513.2021.9562831
Author information
Authors and Affiliations
Corresponding author
Additional information
Translated by I. Moshkin
About this article
Cite this article
Kozlova, L.P., Chyong, D.D. A Terminal Sliding Mode with a Neural Network for an Exoskeleton Electric-Drive System. Russ. Electr. Engin. 94, 186–190 (2023). https://doi.org/10.3103/S1068371223030082
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.3103/S1068371223030082