Abstract
While the human hand is a marvelous example of very fine mechanics and control, its current artificial replacements (especially hand prostheses) suffer from several limitations, in particular poor functionality and poor control interface. The first aspect can be solved by developing a new generation of hand prosthesis. The lack of practical control interfaces, instead, is still an open issue. Among several possibilities, surface electromyographic signals (EMG) are considered an interesting source of information to allow human beings to control robotic artifacts. In this paper, a novel approach for the automatic generation of fuzzy classifiers from raw EMG data through genetic evolution is proposed and tested. The preliminary results show the validity of the proposed approach for the control of a multi-DoFs hand prosthesis.
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call