Early signs of underbone motorcyclist muscle fatigue analysis using s-transform based on surface electromyography signals

Tengku Zawawi, Tengku Nor Shuhada (2024) Early signs of underbone motorcyclist muscle fatigue analysis using s-transform based on surface electromyography signals. Doctoral thesis, Universiti Teknikal Malaysia Melaka.

[img] Text (24 Pages)
Early signs of underbone motorcyclist muscle fatigue analysis using s-transform based on surface electromyography signals.pdf - Submitted Version

Download (331kB)
[img] Text (Full Text)
Early signs of underbone motorcyclist muscle fatigue analysis using s-transform based on surface electromyography signals.pdf - Submitted Version
Restricted to Registered users only

Download (14MB)

Abstract

Seven out of ten deaths on the road in Malaysia were motorcycle users and there are 48% probability that the accident is connected to fatigue. In muscle fatigue for underbone motorcycles, there are unavailable detection types of muscle fatigue with standard feature indicators. Identifying early signs of muscle fatigue is important to avoid injury and accidents from prolonged motorcycle riding. The current method is to analyse muscle fatigue for motorcycles in terms of ergonomic assessment tools in the time domain and frequency domain limited to the rider's specific condition during prolonged rides and automated classification type of muscle fatigue for motorcycles is still not being explored. However, it is crucial to know to avoid muscle injuries and accidents. Therefore, this research aim is to detect the onset of muscle fatigue using time-frequency distribution in riders on motorcycles with automated classification type of muscle fatigue during prolonged riding motorcycle. For this purpose, 24 secondary data respondents of riding underbone motorcycle for prolonged ride have been used for signal pre-processing, signal processing using time domain, frequency domain, and time-frequency domain analysis and classification type of muscle fatigue. Eight muscles of the upper limb side are selected as the most activated involved in riding motorcycles which are the Left and Right Extensor Carpi Radialis, Left and Right Trapezius, Left and Right Erector Spinae, and Left and Right Latissimus Dorsi. The time domain and frequency domain are utilised for determining muscle activation of high intensity which Extensor Carpi Radialis for both regions with the highest value of 11.208 of standard deviation (STD) which features reflect the changes in muscle activation of movement intensity. The second question of the research to find the best method TFD method for detecting and measuring muscle fatigue using time-frequency distribution (TFD) methods either are Spectrogram or S-transform. Five time-frequency representation (TFR) from TFD method features which are instantaneous RMS voltage (Vrms(t)), instantaneous mean frequency (IMNF), instantaneous median frequency (IMDF), instantaneous energy distribution (IED) and instantaneous frequency variance (IFV) used as muscle fatigue indicators for finding muscle fatigue onset and types of muscle fatigue. S-Transform is chosen as the best method of TFD analysis with lower relative error, higher accuracy, and lower computational complexity for 20% for better performance. The automation classification process from ML is applied to five types of classifiers: linear discriminant analysis (LDA), support vector machine (SVM), K-Nearest Neighbour (KNN), artificial neural network (ANN), and Naive Bayes (NB). The results show that ANN offers the highest accuracy 98.8% for classification performance evaluation to automate recognising the pattern of types of muscle fatigue. Therefore, this study concludes S-Transform technique with the proposed muscle fatigue indicators feasible to be apply for muscle fatigue detection and to classify the types of fatigue for awareness of rider motorcycle to be alert. Knowing the early signs of muscle fatigue can prevent a serious degree of muscle fatigue that would increase the risk of musculoskeletal disorders injuries accidents not only among rider motorcycle but also driver for other vehicles.

Item Type: Thesis (Doctoral)
Uncontrolled Keywords: Muscle fatigue, Motorcycle accidents, Ergonomic assessment tools
Divisions: Library > Tesis > FTKE
Depositing User: Muhamad Hafeez Zainudin
Date Deposited: 17 Mar 2025 12:02
Last Modified: 17 Mar 2025 12:02
URI: http://eprints.utem.edu.my/id/eprint/28566
Statistic Details: View Download Statistic

Actions (login required)

View Item View Item