yEMG-Based 1D Stiffness Modeling and Classification in Fingertip Grasping Parmak Ucu Kavramada yEMG Tabanli 1B Sertlik Modellemesi ve Siniflandirilmasi
34th Signal Processing and Communications Applications Conference, SIU 2026, İstanbul, Türkiye, 7 - 10 Temmuz 2026, (Tam Metin Bildiri)
- Yayın Türü: Bildiri / Tam Metin Bildiri
- Doi Numarası: 10.1109/siu71813.2026.11636970
- Basıldığı Şehir: İstanbul
- Basıldığı Ülke: Türkiye
- Anahtar Kelimeler: pinch grasp, prosthetic hand, stiffness modeling, Surface electromyography signal
- İstanbul Medipol Üniversitesi Adresli: Evet
Özet
This study models the mechanical interaction at the fingertip during natural pinch grasping using a control-oriented one-dimensional (1D) stiffness representation. In an experimental setup with different spring stiffness conditions, surface electromyography (sEMG) signals acquired from the Flexor Digitorum Superficialis (FDS) and Extensor Digitorum Communis (Extensor Digitorum Communis) muscles were processed, and root mean square (RMS), average amplitude change (AAC), and co-contraction index (CCI)-based features were extracted. These features were associated with mechanically computed stiffness derived from force and displacement measurements and mapped into three discrete classes: soft, medium, and hard. Results show that the hard condition is clearly separated, gönüllü(subject)specific support vector machine (SVM) models achieve strong accuracy, and limited calibration samples significantly improve personalization. The proposed framework provides an interpretable, low-complexity, and real-time feasible control layer for prosthetic hand and tele-impedance applications.