← All datasets

Full-body wearable sensor suite (IMUs, pressure insoles, EMG) synchronized with lab-grade motion capture and force plates from 10 participants across 33 task variations of simulated manual labor: lead blanket lifting, squatting, wagon push/pull, shoveling, stair and ladder climbing, and level or graded walking. Includes OpenSim inverse kinematics, inverse dynamics, and right knee joint reaction forces as labels for real-time biomechanics estimation.

Details

Subjects

n = 10 · 2F / 8M · 25.2 +/- 3.1 yr (mean +/- SD)

Protocol / load

Lead blanket lifting, weighted and unweighted squats, loaded wagon push/pull, shoveling, stair and ladder climbing

Equipment

Microstrain 6-axis IMUs (back, pelvis, thigh, shank, foot), XSENSOR X4 pressure insoles, EMG on right leg, optical motion capture, force plates, load cells

Sampling

imu 200 Hz · pressure 80 Hz · mocap 200 Hz · force_plate 2000 Hz

Formats

csv, mat

License

CC-BY-4.0

Keywords

exoskeleton-control, deep-learning, joint-moments, opensim, manual-labor, real-time

Added to catalog

2026-07-25

Exoskeleton

Record type

Controller training data

Body region

Back, Knee

Study design

No device worn; wearable-sensor labels for real-time joint-moment estimation aimed at exoskeleton control

This record is part of the exoskeleton collection.

Cite

Cite the dataset as its authors request (check the source page for an official citation); this BibTeX is a convenience starting point.

@misc{gt-second-skin,
  author = {Casey, R. and Nuesslein, C. and Davenport, F. and Wheeler, J. and Mazumdar, A. and Sawicki, G.S. and Young, A.J.},
  title  = {{The Second Skin: Wearable Sensor Suite for Real-Time Biomechanics Tracking}},
  year   = {2025},
  doi    = {10.1109/TBME.2025.3589996},
  url    = {https://repository.gatech.edu/entities/publication/c7a857c8-79d8-4119-8ce0-be568d942f49},
}