Development of an EMG-based hand prosthesis with machine learning

Mohammad Zawawi, Muhammad Muzhafar and Wan Daud, Wan Mohd Bukhari and Al-Nuaimi, Mohammed and Tarmizi, Ahmad Izzuddin (2026) Development of an EMG-based hand prosthesis with machine learning. Przeglad Elektrotechniczny, 1. pp. 164-174. ISSN 0033-2097

Text 0218330012026177462995.pdf Available under License Creative Commons Attribution Non-commercial No Derivatives. Download (3MB)

Abstract

This study describes the design and comparative assessment of an EMG-based hand prosthesis control strategy using three different systems: Artificial Neural Networks (ANN), Fuzzy Logic (FL), and a Threshold-Based (TB) Arduino system as a baseline. Testing of the systems was conducted in accordance with a standardized protocol with ten healthy participants. The testing results showed that the FL system had the highest classification accuracy (91.3±3.5%) and low tracking error (0.11 ± 0.02 rad RMSE); FL also was the most effective system of the three used for handling the complexity and uncertainty of EMG signals with little training data. The ANN system did show the possibility of achieving similar high precision, but, as it experienced low initial classification performance (55.0±5.1%), this suggests that it was limited because of a lack of training data and therefore needs diverse training data in order to realize the full potential of the ANN system. The TB system was the fastest of the three in terms of response latency (150 ± 15ms). Overall, findings from this study suggest that the control strategy should emphasize the type of training data available; with FL being the most robust in limited training data to contest with the complexity of EMG signals, while ANN systems require ample training data for complex control performance.

Item Type:	Article
Uncontrolled Keywords:	EMG-based hand, Fuzzy logic, Machine learning, Smart prosthetic control
Divisions:	Faculty Of Electrical Technology And Engineering
Depositing User:	Norfaradilla Idayu Ab. Ghafar
Date Deposited:	04 Sep 2026 01:26
Last Modified:	04 Sep 2026 01:26
URI:	http://eprints.utem.edu.my/id/eprint/30407
Statistic Details:	View Download Statistic

Actions (login required)

View Item