November 5, 2025

EMG pattern recognition by neural networks for multi fingers control

Key Points

Recognition of finger movement and joint angles was achieved using EMG signals and neural networks, enhancing robotic control.
Key metrics include classification of movements, integrating data from 5 fingers and 10 joints into a robotic hand design.
Assessment involved online simulations and experiments focused on real-time interpretations of EMG patterns.
This implies potential improvements in the design of intuitive robotic systems; further applications may follow.

Abstract

This paper proposes that EMG patterns can be analyzed and classified by neural networks. Through experiments and on-line simulations, it is demonstrated that recognition of not only finger movement but also joint angles in continuously finger movement, based on EMG patterns, can be successfully accomplished.We also demonstrate a EMG controlled 5 fingers 10 joints robot hand.

EMG pattern recognition by neural networks for multi fingers control

Key Points

Abstract

Cite This Study

Also Consider

Also Consider