EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Robot steering-angle prediction lightweight network based non-local attention and lane guidance

Jing Niu iD, Jiahao Zheng, Chuanyan Shen, Guanghao Gao, Guoqiang Song, Shifeng Liu, Yibo Wang

PLOS One · 2026

DOI: 10.1371/journal.pone.0339409 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS One, 2026. Verfasst von Jing Niu und 6 weiteren Autor:innen. Beteiligte Einrichtung: School of Mechatronics and Automotive Engineering, Tianshui Normal University, Tianshui, China. Hauptthemen laut MEDLINE: Autonomous Robots, Robotics. Die Arbeit nennt 44 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Key R&D Program of Gansu Province, Gansu Province Higher Education Industry Support Program Project, Tianshui Normal University Industry Support and Guidance Project.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Predicting the steering angle of robots is a core challenge in autonomous navigation. This paper proposes a novel end-to-end prediction network that integrates non-local attention and lane line guidance mechanisms to significantly reduce computational time and improve prediction accuracy. Built upon the ResNet architecture, the network incorporates a Non Local Block mechanism to enhance global context modeling and a Ghost Module to reduce parameter count and improve feature extraction efficiency. To further optimize training, a ReduceLROnPlateau learning rate scheduler is employed to adaptively adjust the learning rate, effectively mitigating overfitting. Additionally, a lane line annotation method, which combines Canny edge detection with the Hough transform, is used to semantically guide the input images. This enhances the representational power and generalizability of the training data. Experimental results demonstrate that the proposed network outperforms the baseline across multiple evaluation metrics. Under identical experimental conditions, the proposed model achieves a 54.88% increase in inference speed and reduces the mean absolute error (MAE) and root mean square error (RMSE) by 8.47% and 18.23% respectively. Ablation studies further confirm that the combination of the Non-Local Block and Ghost Module significantly improves both expressive capacity and computational efficiency of the model. These findings suggest that the proposed method offers a high-precision, efficient, and low-latency visual perception solution for real-time autonomous navigation of wheeled robots in complex environments.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Autonomous RobotsRoboticsAlgorithmsConvolutional Neural NetworksTransportation

Autor:innen und Institutionen

  • School of Mechatronics and Automotive Engineering, Tianshui Normal University, Tianshui, China.
  • Key Laboratory of Resource Utilization of Agricultural Solid Waste in Gansu Province, Tianshui Normal University, Tianshui, China.

ORCID-Profile: Jing Niu

Förderung

  • Key R&D Program of Gansu Province
  • Gansu Province Higher Education Industry Support Program Project
  • Tianshui Normal University Industry Support and Guidance Project
  • Tianshui Normal University Innovation and Entrepreneurship Project
  • Tianshui Normal University Graduate Innovation Project

Zitieren

APA

Niu, J., Zheng, J., Shen, C., Gao, G., Song, G., Liu, S., & Wang, Y. (2026). Robot steering-angle prediction lightweight network based non-local attention and lane guidance. PLOS One. https://doi.org/10.1371/journal.pone.0339409

Vancouver

Niu J, Zheng J, Shen C, Gao G, Song G, Liu S, et al. Robot steering-angle prediction lightweight network based non-local attention and lane guidance. PLOS One. 2026. doi:10.1371/journal.pone.0339409

BibTeX
@article{Niu2026,
  author = {Niu, Jing and Zheng, Jiahao and Shen, Chuanyan and Gao, Guanghao and Song, Guoqiang and Liu, Shifeng and Wang, Yibo},
  title = {Robot steering-angle prediction lightweight network based non-local attention and lane guidance},
  journal = {PLOS One},
  year = {2026},
  doi = {10.1371/journal.pone.0339409}
}
Kostenlos starten →