Preliminary Validation of a Wearable IoT System for Ergonomic Metrics and Activity Classification

This paper proposes a Wearable Internet of Things-based system for monitoring the well-being of operators working in assembly lines. The system consists of 3 Inertial Measurement Units (IMU) used to recognize the activity and orientation of operators, a board for the acquisition of physiological signals, an Indoor Positioning System (IPS) to track the operator movements in real-time, and a smart Radio Frequency IDentification (RFID) glove able to recognize passive tags on tools and components. A preliminary experimental evaluation was carried out to assess the capability of the system to digitize the European Assembly Worksheet (EAWS) ergonomic scores and to classify the operator's activities using the data from the IMUs by training a Booster Tree algorithm. From the results, it can be highlighted that the system is capable of performing EAWS digitization and is able to classify the operator's activity with an accuracy of 95.83%.