| Titre : | Neuro-Fuzzy Control of a Manipulator Arm by Electro-Myographic (EMG) Signals |
| Auteurs : | HAMZI Maroua, Auteur ; Mohamed Boumehraz, Directeur de thèse |
| Type de document : | Monographie imprimée |
| Editeur : | Université Mohamed Kheider, 2026 |
| Langues: | Anglais |
| Mots-clés: | Myoware muscle sensor, linear discriminant analysis (LDA), wavelet transform, ANFIS, feature extraction. |
| Résumé : |
Electromyography (EMG) is the record of electrical activity generated by skeletal muscles,
providing information on the muscle function and movement. Several signal analysis methods have been developed in the time and frequency domains for use in different engineering applications such as myoelectric control of a prosthesis and movement analysis, mainly to overcome the complexity of EMG signals. The present thesis relies on a Myoware muscle sensor and MPU6050 board to acquire EMG signals from ten healthy volunteers in different forearm positions. Root mean square (RMS), standard deviation (STD), and mean absolute value (MAV) are calculated from each EMG signal and selected as representative features for an LDA and ANFIS classifiers to estimate the forearm flexion angles for the control of a manipulater arm purpose by developing MATLAB tools. Accordingly, this study aims to compare the effectiveness of features calculated from the EMG signal and those calculated from its wavelet decomposition. The experiment’s results demonstrate the proposed method’s efficiecy in estimating forearm angles of flexion using only one channel of EMG signal for four gesture classes. |
| Type de document : | Thése doctorat |
Disponibilité (1)
| Cote | Support | Localisation | Statut | Emplacement | |
|---|---|---|---|---|---|
| Th/1447 | Thèse de doctorat | BIB.FAC.ST. | Empruntable | Magazin |
Erreur sur le template
