An IMU-Based Wearable Device Utilizing Error-State Kalman Filtering for Motion Capture
Applied Sciences · 17.11.2026 · 16 (22) · 11392
DOI: 10.3390/app162211392 ↗Kurz gefasst
Fachartikel in der Zeitschrift Applied Sciences, online erschienen am 17.11.2026. Verfasst von Liang Shan und 1 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Qingdao University.
Abstract (Originalsprache)
Inertial measurement units (IMUs) provide a portable and infrastructure-free solution for human motion capture, but accumulated integration errors lead to substantial drift during continuous motion. This paper presents a wearable device for human motion capture using wireless IMU sensing nodes and a constraint-aided error-state Kalman filtering framework. For lower-limb motion capture, three IMU nodes are placed on the pelvis and bilateral ankles to reduce the number of body-worn sensors. Zero-velocity updates are applied during detected stance phases, and biomechanical constraints are introduced to suppress inertial drift and improve motion consistency. Closed-loop walking experiments and Vicon-referenced evaluation demonstrate that the proposed device achieves a relative trajectory drift of 2.64% and an overall root-mean-square error (RMSE) of 15.334 mm, together with improved biomechanical consistency. The proposed device provides a practical solution for wearable motion capture with sparse sensor deployment.
Förderung
- Qingdao University
Zitieren
APA
Shan, L., & Liu, Y. (2026). An IMU-Based Wearable Device Utilizing Error-State Kalman Filtering for Motion Capture. Applied Sciences, 16(22), 11392. https://doi.org/10.3390/app162211392
Vancouver
Shan L, Liu Y. An IMU-Based Wearable Device Utilizing Error-State Kalman Filtering for Motion Capture. Applied Sciences. 2026;16(22):11392. doi:10.3390/app162211392
BibTeX
@article{Shan2026,
author = {Shan, Liang and Liu, Yinhua},
title = {An IMU-Based Wearable Device Utilizing Error-State Kalman Filtering for Motion Capture},
journal = {Applied Sciences},
year = {2026},
volume = {16},
number = {22},
pages = {11392},
doi = {10.3390/app162211392}
}
EUVIMED