Projects:2017s1-158 Electromyographic Signal Processing for Controlling an Exoskeleton

From Projects
Revision as of 12:48, 22 September 2017 by A1637416 (talk | contribs)
Jump to: navigation, search

Introduction

EMG signals reflect bioelectrical activities of muscles, which means that physical condition can be shown by the EMG signal. Accordingly, it is important to recognize EMG signal features accurately in not only areas of exoskeleton but also areas of healthcare. The project is to design a feasible and efficient algorithm to process the EMG signal which is extracted on the body surface of lower limbs, thereby controlling the exoskeleton.Surface electrodes and a MyoWare Muscle Sensor are selected to pick up the EMG signal, and an Arduino 2560 Mega board is used to convert the analog signal to digital signal. We use the Xbee module to obtain the digital EMG signal on PC and after that Matlab is used to do the FFT to the signal. Significant features are extracted from the spectrogram and ANN is used to classify the features to different motion patterns.


Project Team

Kaitai Li

Ravinder Singh Sandhu


Supervisors

Mathias Baumert

Tien-Fu Lu

Retrieved from "https://projectswiki.eleceng.adelaide.edu.au/projects/index.php?title=Projects:2017s1-158_Electromyographic_Signal_Processing_for_Controlling_an_Exoskeleton&oldid=8678"