Worker Activity Recognition in Smart Manufacturing Using IMU and sEMG Signals with Convolutional Neural Networks

Abstract

In a smart manufacturing system involving workers, recognition of the worker's activity can be used for quantification and evaluation of the worker's performance, as well as to provide onsite instructions with augmented reality. In this paper, we propose a method for activity recognition using Inertial Measurement Unit (IMU) and surface electromyography (sEMG) signals obtained from a Myo armband. The raw 10-channel IMU signals are stacked to form a signal image. This image is transformed into an activity image by applying Discrete Fourier Transformation (DFT) and then fed into a Convolutional Neural Network (CNN) for feature extraction, resulting in a high-level feature vector. Another feature vector representing the level of muscle activation is evaluated with the raw 8-channel sEMG signals. Then these two vectors are concatenated and used for work activity classification. A worker activity dataset is established, which at present contains 6 common activities in assembly tasks, i.e., grab tool/part, hammer nail, use power-screwdriver, rest arm, turn screwdriver, and use wrench. The developed CNN model is evaluated on this dataset and achieves 98% and 87% recognition accuracy in the half-half and leave-one-out experiments, respectively.