LE QUY DON
Technical University
VietnameseClear Cookie - decide language by browser settings

Self-Tuning PID Controller Using a Neural Network for Nonlinear Exoskeleton System

Belov, M.P. and Truong, D.D. and Van Tuan, P. (2021) Self-Tuning PID Controller Using a Neural Network for Nonlinear Exoskeleton System. In: 2nd International Conference on Neural Networks and Neurotechnologies, NeuroNT 2021, 6/16/2021.

Full text not available from this repository. (Upload)

Abstract

The article proposes the application of method analog neural networks (NN) with a radial basis function (RBF) and self-tuning of the proportional-integral derivative (PID) controller coefficients for a nonlinear control system of the lower extremities of the exoskeleton. The implementation of accurate and high-quality control of nonlinear systems, including parameter uncertainties and external disturbances, is possible using an analog neuron of a network controller, which has the ability to continuously learn and adapt. The NN in the PID controller allows us to correct errors that arise due to uncertainties and changes in parameters during the movement of the lower extremities of the exoskeleton. The effect of the proposed control algorithm is demonstrated by means of simulation in the Matlab/Simulink environment. © 2021 IEEE.

Item Type: Conference or Workshop Item (Paper)
Divisions: Faculties > Faculty of Control Engineering
Identification Number: 10.1109/NeuroNT53022.2021.9472852
Uncontrolled Keywords: Controllers; Electric control equipment; Exoskeleton (Robotics); MATLAB; Neurons; Proportional control systems; Quality control; Three term control systems; Two term control systems; Analog neural network; Control of nonlinear system; Exoskeleton systems; External disturbances; MATLAB/Simulink environment; Parameter uncertainty; Proportional integral derivative controllers; Radial Basis Function(RBF); Neural networks
Additional Information: Conference code: 170810. Language of original document: English.
URI: http://eprints.lqdtu.edu.vn/id/eprint/8621

Actions (login required)

View Item
View Item