The WGD—A Dataset of Assembly Line Working Gestures for Ergonomic Analysis and Work-Related Injuries Prevention

Abstract

This paper wants to stress the importance of human movement monitoring to prevent musculoskeletal disorders by proposing the WGD—Working Gesture Dataset, a publicly available dataset of assembly line working gestures that aims to be used for worker’s kinematic analysis. It contains kinematic data acquired from healthy subjects performing assembly line working activities using an optoelectronic motion capture system. The acquired data were used to extract quantitative indicators to assess how the working tasks were performed and to detect useful information to estimate the exposure to the factors that may contribute to the onset of musculoskeletal disorders. The obtained results demonstrate that the proposed indicators can be exploited to early detect incorrect gestures and postures and, consequently to prevent work-related disorders. The approach is general and independent on the adopted motion analysis system. It wants to provide indications for safely performing working activities. For example, the proposed WGD can also be used to evaluate the kinematics of workers in real working environments thanks to the adoption of unobtrusive measuring systems, such as wearable sensors through the extracted indicators and thresholds.

A Classification Method for Workers’ Physical Risk

“…About 84% of all non-fatal injuries and illnesses leading to days away from work in 2020 involved slips, trips, falls, overexertion or exposure to harmful substances or environmental conditions, according to International Labour Organisation (ILO) statistics [ 1 , 2 ]. With the advent of the Industry 4.0 paradigm, worker monitoring to understand their state during the working day is crucial in order to prevent and identify risk factors before they can lead to actual damage to health [ 3 ]. Among all the aforementioned injury causes, the phenomenon of falling is the one that is most analyzed in the literature.…”

Robot Learning by Demonstration with Dynamic Parameterization of the Orientation: An Application to Agricultural Activities

“…The attractor landscape allows for replications of the recorded trajectory by means of a weighted sum of equally spaced Gaussian Kernels. A generic modeling approach to learning the landscape attractor is proposed in [25] and consists of extracting the weight parameters (DMP parameters) from demonstrated movements, in a single shot, by means of linear regression algorithms [26].…”

A Semiautonomous Control Strategy Based on Computer Vision for a Hand–Wrist Prosthesis

“…The loss of an upper limb has profound physical, psychological, and social consequences [1]. It significantly impacts an individual's ability to perform activities of daily living and work-related tasks [2], limiting their independence and overall quality of life. Despite advancements in prosthetic technology, statistics show that less than half of amputees (44.7%) use their prosthetic device for more than eight hours per day.…”