Design and Validation of a Morphing Myoelectric Hand Posture Controller Based on Principal Component Analysis of Human Grasping

This paper describes the design and validation of a morphing myoelectric hand controller based on principal component analysis of human grasping

Jacob L. Segil; Richard F. ff. Weir


Scholarcy highlights

  • An ideal myoelectric prosthetic hand should have the ability to continuously morph between postures like an anatomical hand
  • Subjects had to co-contract in order to reach regions of the principal components domain not close to an EMG axis when using myoelectric control
  • Our study verifies that a myoelectric controller based on principal components of human grasping can control a multi-multifunctional virtual hand in a continuously morphing fashion
  • A validation experiment studied the performance of the controller using clinically practiced techniques including myoelectric control site selection, commercially available surface electrodes, and standard EMG filtering
  • A correlation analysis was performed in order to understand the relationship between distance in the PC domain and performance
  • Santello et al found that the first two PCs described over 80% of the variance in hand posture
  • The experimental results presented indicate that the controller based on PCA of human grasping provides an effective method for nonamputee subjects to morph a high DOM virtual hand into functional grasps

Need more features? Save interactive summary cards to your Scholarcy Library.