Home > Published Issues > 2021 > Volume 10, No. 8, August 2021 >

Non-Linear Motorized Prosthetic Hand System with Gradient Descent Tuning Technique

Mohd Hafiz Jali 1, Rozaimi Ghazali 1, Chong Chee Soon 1, and Ahmad Razif Muhammad 2
1. Centre for Robotics and Industrial Automation, Faculty of Electrical Engineering, Universiti Teknikal Malaysia Melaka, Hang Tuah Jaya, 76100, Durian Tunggal, Melaka, Malaysia
2. Faculty of Applied Sciences, Ton Duc Thang University, Ho Chi Minh City, Vietnam

Abstract— This paper describes the controller design for a nonlinear motorized prosthetic finger system. This system can be used as a human assistive device for amputee. Since the prosthetic device is worn by human, the accuracy of the system is crucial to avoid unnecessary injury. In addition, the mathematical modelling of the system needs to be developed appropriately to ensure the accuracy of the system. Various types of controllers can be used to obtain a stable nonlinear actuated finger system, such as Proportional Integral (PI), Proportional Integral and Derivative (PID), and Fuzzy Logic controllers. In this work, the Proportional, Integral, and Derivative (PID) controller will be used. The tuning of the PID control parameter is for positioning feedback control of the motor. To improve the transient response performance of the motor, Gradient Descent and Auto--Tuning techniques have been used to obtain the parameters of the PID controller. Comparison between these techniques and the comparison with the previous work is carried out. It is observed from the results, Gradient Descent tuning technique outperforms the Auto-Tuning technique.

Index Terms— prosthetic finger, PID controller, Ziegler-Nichols, gradient descent, tuning technique, transient response

Cite: Mohd Hafiz Jali, Rozaimi Ghazali, Chong Chee Soon, and Ahmad Razif Muhammad, "Non-Linear Motorized Prosthetic Hand System with Gradient Descent Tuning Technique," International Journal of Mechanical Engineering and Robotics Research, Vol. 10, No. 8, pp.429-435, August 2021. DOI: 10.18178/ijmerr.10.8.429-435

Copyright © 2021 by the authors. This is an open access article distributed under the Creative Commons Attribution License (CC BY-NC-ND 4.0), which permits use, distribution and reproduction in any medium, provided that the article is properly cited, the use is non-commercial and no modifications or adaptations are made.