Enhancing Trajectory Estimation With Convolutional IMU Transformer
IEEE Sensors Journal · 01.05.2026 · 26 (9) · 14207-14218
DOI: 10.1109/jsen.2026.3673719 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Sensors Journal, erschienen am 01.05.2026. Verfasst von Sunghyun Kim und 5 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 40 Literaturangaben. Gefördert von: National Research Foundation of Korea (NRF) through Korean Government [Ministry of Science and Information and Communication Technology (MSIT)], Korea Evaluation Institute of Industrial Technology (KEIT) through the Ministry of Trade, Industry, and Energy, Education and Research Promotion Program of Korea University of Technology and Education (KOREATECH) in 2025.
Abstract (Originalsprache)
Accurate trajectory estimation using low-cost inertial measurement unit (IMU) data remains a challenging task due to sensor noise, bias, and drift. In this article, we propose the convolutional IMU Transformer (CoIT), a deep neural network that combines convolutional token embedding with a hierarchical Transformer architecture to effectively capture both local motion features and global temporal dependencies from raw IMU measurements. The proposed network is designed to overcome the limitations of traditional dead reckoning methods and prior learning-based approaches that rely heavily on inaccurate orientation estimates. CoIT further incorporates a multistage convolutional projection and downsampling strategy to improve computational efficiency while preserving representational capacity. We evaluate CoIT on three large-scale public datasets (RIDI, RoNIN, and NeurIT) and demonstrate consistent performance gains over existing state-of-the-art models. On the RoNIN dataset, CoIT achieves an absolute trajectory error (ATE) of 3.39 ± 0.59 m, a time-normalized relative trajectory error (T-RTE) of 2.63 ± 0.57 m, and a distance-normalized relative trajectory error (D-RTE) of 0.17 ± 0.02 m while also demonstrating robust performance on unseen subjects and competitive cross-domain transfer from pedestrian motion (RoNIN) to indoor robot motion (NeurIT) without finetuning. Furthermore, an ablation study on model complexity reveals a clear tradeoff between accuracy and latency, demonstrating the architecture’s scalability for practical deployment. These results highlight the robustness and statistical reliability of the proposed model within the evaluated public benchmarks in GPS-denied settings.
Förderung
- National Research Foundation of Korea (NRF) through Korean Government [Ministry of Science and Information and Communication Technology (MSIT)]
- Korea Evaluation Institute of Industrial Technology (KEIT) through the Ministry of Trade, Industry, and Energy
- Education and Research Promotion Program of Korea University of Technology and Education (KOREATECH) in 2025
Zitieren
APA
Kim, S., Noh, S., Jeong, S., Kim, C., Lee, H., & Kim, J. (2026). Enhancing Trajectory Estimation With Convolutional IMU Transformer. IEEE Sensors Journal, 26(9), 14207-14218. https://doi.org/10.1109/jsen.2026.3673719
Vancouver
Kim S, Noh S, Jeong S, Kim C, Lee H, Kim J. Enhancing Trajectory Estimation With Convolutional IMU Transformer. IEEE Sensors Journal. 2026;26(9):14207-14218. doi:10.1109/jsen.2026.3673719
BibTeX
@article{Kim2026,
author = {Kim, Sunghyun and Noh, Sooyoung and Jeong, Seungik and Kim, Changhyun and Lee, Haeyun and Kim, Jihun},
title = {Enhancing Trajectory Estimation With Convolutional IMU Transformer},
journal = {IEEE Sensors Journal},
year = {2026},
volume = {26},
number = {9},
pages = {14207--14218},
doi = {10.1109/jsen.2026.3673719}
}
EUVIMED