Wearable IMU Gait Data for Physics-Informed Fatigue Assessment

A 602.1 KB dataset supporting a physics-informed transformer model for gait fatigue assessment. The data, authored by Kai Ding and released under CC-BY-4.0, was last updated on April 13, 2026. It contains gait cycles from multiple participants, collected using wearable inertial measurement units (IMUs).

Use Cases

• Training models for multi-level fatigue classification based on gait cycle data.
• Developing sensor fusion methods based on distributed IMU measurements.
• Implementing physics-informed learning with biomechanical constraints like kinematic limits and bilateral symmetry.
• Testing model robustness under conditions of sensor noise or individual sensor failure.
• Conducting cross-subject and cross-environment validation for field deployment.

Strengths

• Dataset is small (602.1 KB), facilitating quick download and exploration.
• Includes cross-subject and cross-environment validation, suggesting tested generalization capability.
• Released under the permissive CC-BY-4.0 license, allowing broad reuse.

Limitations

• Row count and column-level documentation are unknown, which may limit suitability assessment.
• Data scale is limited; the 602.1 KB size suggests a small sample, potentially insufficient for large-scale model training.

Provenance

Source

Kai Ding via figshare.

Collection Method

Gait cycles collected from multiple participants using wearable inertial measurement units (IMUs).

Time Range

null

Freshness

Last updated 2026-04-13 05:30:59; freshness should be verified.

Geography

null