Thumb-tip force prediction based on hill’s muscle model using electromyogram and ultrasound signal

• Published in: International Journal of Computational Intelligence Systems
• Main Author: Sidek S.N.; Roslan M.R.; Sidek S.; Khalid M.S.M.
• Format: Article
• Language: English
• Published: Atlantis Press 2018
• Online Access: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85045612530&doi=10.2991%2fijcis.11.1.18&partnerID=40&md5=df0f6f9a8b2cd57bee8f001eb1c940e1

The use of prostheses is necessary to restore lost limbs to a level of functionality to enable activity of daily living. Many prostheses are now using myoelectric based control techniques to operate. However, to develop a model based controller for the system remains a challenge as accurate model is necessary. This study investigates the use of electromyogram and ultrasound signal to predict thumb tip force based on Hill’s Muscle model. The results obtained has shown a significant improvement in the prediction of thumb tip force as much as 31.45% of average RMSE over the benchmark model that leverages on biomechanics model and active marker to characterize the muscle. © 2018, the Authors.