FR-PDR: Factor Graph and Residual Attention Neural Network Improved Pedestrian Dead Reckoning With Shoulder-Mounted IMU
IEEE Sensors Journal · 15.05.2026 · 26 (10) · 15500-15512
DOI: 10.1109/jsen.2026.3677501 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Sensors Journal, erschienen am 15.05.2026. Verfasst von Ziwei Yue und 4 weiteren Autor:innen. Gefördert von: Beijing Natural Science Foundation, National Key Research and Development Program of China, Hong Kong Scholars Program.
Abstract (Originalsprache)
The pedestrian dead reckoning (PDR) plays an increasingly important role in the personnel positioning of the industrial Internet of Things (IoT). However, PDR based on shoulder-mounted inertial measurement units (IMUs) faces issues such as the divergence of heading errors over time and inaccurate step length (SL) estimation. To address these challenges, this article proposed FR-PDR, which is improved by a factor graph and residual attention neural network (RANN) with a shoulder-mounted IMU. On the one hand, factor graph optimization (FGO) combined with quasi-static magnetic constraint (QSMC) integrates attitudes based on magnetic field and inertia, effectively correcting the cumulative error of gyroscopes. On the other hand, RANN constructs an accurate SL estimation model under different walking speeds using only sparse training datasets. Through experimental verification, the method proposed has reduced the cumulative heading error to 0.24°/min and the singlestep mean absolute error to 0.04 m. The collaborative optimization of these two errors ultimately reduces the positioning error to 0.81%. The experimental results indicate that the method proposed significantly improves the positioning accuracy of PDR system with a shoulder-mounted IMU, greatly enhancing pedestrian indoor navigation performance.
Förderung
- Beijing Natural Science Foundation
- National Key Research and Development Program of China
- Hong Kong Scholars Program
- China Postdoctoral Science Foundation
- Fundamental Research Funds for the Central Universities
Zitieren
APA
Yue, Z., Wang, J., Xia, M., Yang, D., & Chen, W. (2026). FR-PDR: Factor Graph and Residual Attention Neural Network Improved Pedestrian Dead Reckoning With Shoulder-Mounted IMU. IEEE Sensors Journal, 26(10), 15500-15512. https://doi.org/10.1109/jsen.2026.3677501
Vancouver
Yue Z, Wang J, Xia M, Yang D, Chen W. FR-PDR: Factor Graph and Residual Attention Neural Network Improved Pedestrian Dead Reckoning With Shoulder-Mounted IMU. IEEE Sensors Journal. 2026;26(10):15500-15512. doi:10.1109/jsen.2026.3677501
BibTeX
@article{Yue2026,
author = {Yue, Ziwei and Wang, Jiale and Xia, Ming and Yang, Dongkai and Chen, Wu},
title = {FR-PDR: Factor Graph and Residual Attention Neural Network Improved Pedestrian Dead Reckoning With Shoulder-Mounted IMU},
journal = {IEEE Sensors Journal},
year = {2026},
volume = {26},
number = {10},
pages = {15500--15512},
doi = {10.1109/jsen.2026.3677501}
}
EUVIMED