EUVIMEDEUVIMED entdecken
Crossref

Physics-Guided Deep Neural Networks for Radar-Based UAV Recognition in Different Environments With No Prior In Situ Data

Sean Kearney, Sevgi Z. Gurbuz

IEEE Transactions on Aerospace and Electronic Systems · 2026 · 62 · 9875-9891

DOI: 10.1109/taes.2026.3685229 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Aerospace and Electronic Systems, 2026. Verfasst von Sean Kearney und 1 weiteren Autor:innen. Die Arbeit nennt 62 Literaturangaben. Gefördert von: Air Force Office of Scientific Research.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Classifying uncrewed aerial vehicle (UAV) radar micro-Doppler signatures with deep neural networks (DNNs) suffers a degradation in performance when there is a difference in the distribution of training and test data due to the differences in clutter backscatter and signal-to-interference-plus-noise ratio (SINR) of different environments. This is further exacerbated in low-SINR samples in which the helicopter motor rotation (HERM) line characteristic of UAV blade rotations becomes only faintly visible. As HERM lines are a critical feature enabling the recognition and discrimination of various types of UAVs, this article addresses the twofold challenge of 1) how to accurately synthesize HERM lines in training data for UAV classification and 2) how to enhance recognition performance in previously unseen different environments, especially when the SINR is low. Toward these aims, we propose a physics-guided approach in which a semantic segmentation network trained only using low-fidelity computer aided design (CAD)-models of UAVs is integrated into the discriminator of a generative adversarial network (GAN) for training data synthesis, and on the front end of a DNN for classification. Our results show that physics-guided semantic segmentation improves the representation of HERM lines in GAN-synthesized samples and enables improved classification of low-SINR samples exhibiting HERM lines when no prior in situ data at that SINR are used during training. Consequently, our approach provides a way of enhancing UAV recognition performance in different operational environments, from which it is often difficult to obtain training data prior to deployment.

Förderung

  • Air Force Office of Scientific Research

Zitieren

APA

Kearney, S., & Gurbuz, S. Z. (2026). Physics-Guided Deep Neural Networks for Radar-Based UAV Recognition in Different Environments With No Prior In Situ Data. IEEE Transactions on Aerospace and Electronic Systems, 62, 9875-9891. https://doi.org/10.1109/taes.2026.3685229

Vancouver

Kearney S, Gurbuz SZ. Physics-Guided Deep Neural Networks for Radar-Based UAV Recognition in Different Environments With No Prior In Situ Data. IEEE Transactions on Aerospace and Electronic Systems. 2026;62:9875-9891. doi:10.1109/taes.2026.3685229

BibTeX
@article{Kearney2026,
  author = {Kearney, Sean and Gurbuz, Sevgi Z.},
  title = {Physics-Guided Deep Neural Networks for Radar-Based UAV Recognition in Different Environments With No Prior In Situ Data},
  journal = {IEEE Transactions on Aerospace and Electronic Systems},
  year = {2026},
  volume = {62},
  pages = {9875--9891},
  doi = {10.1109/taes.2026.3685229}
}
Kostenlos starten →