STA-UNet: Lightweight Spectral Segmentation for Mowing Robots
IEEE Robotics and Automation Letters · 2026-05 · 11 (5) · 5574-5581
DOI: 10.1109/lra.2026.3673901 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Robotics and Automation Letters, 2026-05. Verfasst von Zhiwei Deng und 3 weiteren Autor:innen. Die Arbeit nennt 25 Literaturangaben. Gefördert von: National Key Research and Development Program of China.
Abstract (Originalsprache)
Achieving robust semantic segmentation for autonomous mowing robots requires discriminating between materials with subtle visual differences under challenging outdoor conditions. To address this, we present a novel framework that co-designs feature-specific spectral imaging and a lightweight deep learning network. On the sensing front, we develop a low-cost multispectral camera using a custom multi-band-pass (MBP) filter, engineered to capture distinctive spectral signatures of key classes like grass and shrubs, enabling parallax-free and compact data acquisition. For efficient inference, we propose STA-UNet, a lightweight encoder-decoder network that innovatively incorporates a Spectral Attention (SA) module to adaptively fuse the acquired spectral bands, and a Cross-Scale Texture Perception (CSTP) module to enhance spatial feature representation. Trained on Grass8K—a new 8,191-image multispectral dataset for lawn environments—STA-UNet achieves a remarkable mIoU of 88.63%, outperforming a standard U-Net by 7.11% while utilizing only one-quarter of the parameters. The system achieves real-time segmentation on an edge device with 5 TOPS of computing power, demonstrating a practical, high-performance perception solution for resource-constrained robotic applications.
Förderung
- National Key Research and Development Program of China
Zitieren
APA
Deng, Z., Lv, T., Shi, Z., & Cao, X. (2026). STA-UNet: Lightweight Spectral Segmentation for Mowing Robots. IEEE Robotics and Automation Letters, 11(5), 5574-5581. https://doi.org/10.1109/lra.2026.3673901
Vancouver
Deng Z, Lv T, Shi Z, Cao X. STA-UNet: Lightweight Spectral Segmentation for Mowing Robots. IEEE Robotics and Automation Letters. 2026;11(5):5574-5581. doi:10.1109/lra.2026.3673901
BibTeX
@article{Deng2026,
author = {Deng, Zhiwei and Lv, Tao and Shi, Zhan and Cao, Xun},
title = {STA-UNet: Lightweight Spectral Segmentation for Mowing Robots},
journal = {IEEE Robotics and Automation Letters},
year = {2026},
volume = {11},
number = {5},
pages = {5574--5581},
doi = {10.1109/lra.2026.3673901}
}
EUVIMED