Digital Twin/MARS‐CycleGAN: Enhancing Sim‐to‐Real Crop/Row Detection for MARS Phenotyping Robot Using Synthetic Images
Journal of Field Robotics · 2025-05 · 42 (3) · 625-640
DOI: 10.1002/rob.22473 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Field Robotics, online erschienen am 10.12.2024. Verfasst von David Liu und 3 weiteren Autor:innen. Die Arbeit wurde bisher 16-mal zitiert (Crossref) und nennt 80 Literaturangaben.
Abstract (Originalsprache)
Robotic crop phenotyping has emerged as a key technology for assessing crops' phenotypic traits at scale, which is essential for developing new crop varieties with the aim of increasing productivity and adapting to the changing climate. However, developing and deploying crop phenotyping robots faces many challenges, such as complex and variable crop shapes that complicate robotic object detection, dynamic and unstructured environments that confound robotic control, and real‐time computing and managing big data that challenge robotic hardware/software. This work specifically addresses the first challenge by proposing a novel Digital Twin(DT)/MARS‐CycleGAN model for image augmentation to improve our Modular Agricultural Robotic System (MARS)'s crop object detection from complex and variable backgrounds. The core idea is that in addition to the cycle consistency losses in the CycleGAN model, we designed and enforced a new DT/MARS loss in the deep learning model to penalize the inconsistency between real crop images captured by MARS and synthesized images generated by DT/MARS‐CycleGAN. Therefore, the synthesized crop images closely mimic real images in terms of realism, and they are employed to fine‐tune object detectors such as YOLOv8. Extensive experiments demonstrate that the new DT/MARS‐CycleGAN framework significantly boosts crop/row detection performance for MARS, contributing to the field of robotic crop phenotyping. We release our code and data to the research community ( https://github.com/UGA-BSAIL/DT-MARS-CycleGAN ).
Zitieren
APA
Liu, D., Li, Z., Wu, Z., & Li, C. (2025). Digital Twin/MARS‐CycleGAN: Enhancing Sim‐to‐Real Crop/Row Detection for MARS Phenotyping Robot Using Synthetic Images. Journal of Field Robotics, 42(3), 625-640. https://doi.org/10.1002/rob.22473
Vancouver
Liu D, Li Z, Wu Z, Li C. Digital Twin/MARS‐CycleGAN: Enhancing Sim‐to‐Real Crop/Row Detection for MARS Phenotyping Robot Using Synthetic Images. Journal of Field Robotics. 2025;42(3):625-640. doi:10.1002/rob.22473
BibTeX
@article{Liu2025,
author = {Liu, David and Li, Zhengkun and Wu, Zihao and Li, Changying},
title = {Digital Twin/MARS‐CycleGAN: Enhancing Sim‐to‐Real Crop/Row Detection for MARS Phenotyping Robot Using Synthetic Images},
journal = {Journal of Field Robotics},
year = {2025},
volume = {42},
number = {3},
pages = {625--640},
doi = {10.1002/rob.22473}
}
EUVIMED