FURI | Spring 2018

A Deep Learning Autoencoder for EMG Changepoint Recognition in Robotic Applications

Health icon, disabled. A red heart with a cardiac rhythm running through it.

Electromyography (EMG) sensors can be added to a variety of devices for detecting nerve signals in muscles, but there is a need for better signal analysis techniques and more flexible models for interpreting gestures. Using a machine learning technique called deep learning, the researcher developed a model for detecting gesture changes in EMG data with a high level of accuracy. This kind of model could be used for allowing users to control prosthetics or assistive exoskeletons with a high degree of precision. Future work should be done to develop even better models and make these medical devices more effective.

Student researcher

Portrait of Nester, Elliot

Elliot Nester

Computer systems engineering

Hometown: Tempe, Arizona

Graduation date: Fall 2020