Multi-modal Human Fatigue Classification using Wearable Sensors for Human-Robot Teams

Our main objective of this study is to create a fatigue detection model using real-time data by using wearable sensors. The purpose of this research is to learn more about the way humans experience fatigue in a supervisory human-machine environment. The goal of this study is to evaluate machine learning algorithms that assess fatigue detection and to use robots for adapting its interactions. The environment itself consists of two different tasks to analyze Physical fatigue and Mental fatigue in two different task environments that are (i) Jigsaw puzzle-solving task, and (ii) Pick and Place task. Physical fatigue and mental fatigue are detected using wearable sensors: MYO armband and BioPac Bioharness. During the experiment, the Physiological metrics used are Heart rate, respiration rate, Heart rate variability, posture, breathing wave amplitude, and EMG. All these Physiological signals are collected simultaneously in a real-time task environment. The data collected by these physiological signals are then processed and machine learning and deep learning algorithms are used for further process in building a fatigue detection model