ENTRAPon Dataset

ENTRAPon Dataset

This dataset was created as part of the ENTRAPon project, funded by the German Social Accident Insurance (DGUV) and performed by a consortium consisting of the Ruhr University Bochum, the London South Bank University, the University of Applied Sciences Koblenz and the Institute of Occupational Safety and Health of the German Social Accident Insurance (IFA). Due to the prevalence of slip, trip and fall (STF) accidents in occupational settings, the project was aimed at improving prevention strategies to reduce the number of workplace accidents. During ENTRAPon, subjects were initially exposed to virtual or mechanical perturbations in order to study whether the body learns to reduce the severity of falls. To test the severity of falls subjects walked along a perturbation parcours. This dataset was recorded during the parcours trials and enabled research in the area or machine learning based near-fall detection outside of the ENTRAPon project. During the parcour trials data was collected from 110 participants drawn from two occupational groups (parcel delivery drivers and steelworkers). Participants were instructed to walk along an approximately 15 m long perturbation parcours featuring randomized perturbation elements inducing unpredictable and involuntary near-fall events. All trials were recorded using an Xsens Link system equipped with 17 inertial measurement units (IMUs) at a sampling rate of 120 Hz.

The dataset is designed to support research on:

- Near-fall and fall detection- Human gait stability and perturbation responses- Wearable sensor-based safety systems- Machine learning methods for occupational biomechanics

___

If you use this dataset in your work, please cite the following publication:

Ellegast, R., Schneider, M., Hartmann, U., Kluge, A., Karamanidis, K., Weber, A. Kaufmann, M., et al. (2026) ENTRAPon Dataset (Version 1.0) [Data set]. German Social Accident Insurance (DGUV)URL: https://bitbucket.org/dguv/entrapon-dataset/src/main/DOI: 10.5281/zenodo.18847450

___

Contributions by Authors to the Generation of the ENTRAPon Dataset:

AK, KK, UH, AW and RE developed the study design for the Entrapon studyKK, AW, UH, and RE designed the kinematic data acquisition required for the detection of near-slip, trip, and mistep situations.RE introduced the kinematic measurement approach with IMU sensors with the aim of detecting test/comparison data sets for the development of AI-based measurement systems to detect near-slip, -trip, and -mistep situations at workplacesUH, AW, and MK developed the perturbation course, which was used to simulate realistic near-slip, -trip, and -misstep situations in occupational settingsAW and MK conducted the kinematic measurements with all subjects and prepared the ENTRAPON DatasetMS compiled and tested the ENTRAPON Dataset in its present form

___

Associated Institutes and Laboratories

Institution: Institute for Occupational Safety and Health of the German Social Accident Insurance (IFA)Department: Data Engineering and Artificial Intelligence Key Person: Rolf Ellegast ([e-mail])Website: https://www.dguv.de/ifa

Institution: Hochschule Koblenz – RheinAhrCampusDepartment: Institute for Medical Engineering and Information Processing (MIT)Laboratory: Laboratory for AI‑Based Gait AnalysisKey Person: Ulrich Hartmann Website: https://www.hs-koblenz.de/mit/sportmedizinische-technik/forschung-projekte/labore/labor-fuer-ki-basierte-ganganalyse

Institution: London South Bank UniversityDepartment: School of Applied Sciences - Division of Sport and Exercise Science Research GroupKey Person: Kiros KaramanidisWebsite: https://www.lsbu.ac.uk/research/centres-groups/sport-exercise-science

Institution: Ruhr-Universität BochumDepartment: Lehrstuhl Arbeits-, Organisations- & WirtschaftspsychologieKey Person: Annette KlugeWebsite: https://www.aow.ruhr-uni-bochum.de/