Teleoperated robotic arm movement using electromyography signal with wearable Myo armband

Abstract

The primary purpose of this research is to move a 5-DoF Aideepen ROT3U robotic arm in real-time based on the surface Electromyography (sEMG) signal obtained from a wireless Myo gesture armband to distinguish seven hand movements. The pattern recognition system is employed to analyze these gestures and consists of three main parts: segmentation, feature extraction, and classification. Overlap technique is chosen for segmenting portion of the signal. Six-time domain features, namely, Mean Absolute Value (MAV), Waveform Length (WL), Root Mean Square (RMS), Autoregressive Coefficients (AR), Zero Crossings (ZC), Slope Sign Changes (SSC) are extracted from each segment. While the Support Vector Machines (SVM), Linear Discriminant Analysis (LDA), and K-Nearest Neighbor (K-NN) classifiers are employed in the classification of the seven hand movements. Moreover, a comparison between their performance is carried out to obtain optimum accuracy. The proposed system is tested on datasets extracted from six healthy subjects and the results showed that the SVM achieved higher system accuracy with 95.26% compared to LDA with an accuracy of 92.58%, and 86.41% accuracy achieved by K-NN.