EUVIMEDEUVIMED entdecken
Crossref

Research on excavator environment perception and localization system based on multi-sensor data fusion

Ziwei Wang, Guangjun Liu

Measurement Science and Technology · 29.05.2026 · 37 (22) · 226304

DOI: 10.1088/1361-6501/ae65cc ↗

Kurz gefasst

Fachartikel in der Zeitschrift Measurement Science and Technology, online erschienen am 02.06.2026. Verfasst von Ziwei Wang und 1 weiteren Autor:innen. Die Arbeit nennt 29 Literaturangaben. Gefördert von: Fundamental Research Funds for the Central Universities, Special Funds of the Tongji University for “Sino-German Cooperation 2.0 Strategy”.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Aiming at the limitations of simultaneous localization and mapping (SLAM) algorithms in the application of intelligent unmanned excavator in engineering construction, this paper proposes a multi-sensor fusion SLAM framework integrating LiDAR, camera, and inertial measurement unit (IMU), and designs an excavator environment perception and localization system fusion of LiDAR-inertia system and visual-inertia system (LIS-VIS-Fusion). By tightly coupling IMU data with LiDAR and camera respectively, a LIS and a VIS are formed. In the process of fusing data from the two subsystems, firstly, LiDAR point cloud is utilized to assist visual initialization, achieving accurate depth registration for images. Simultaneously, a visual loop closure detection triggers LiDAR loop closure verification, enhancing the efficiency and accuracy of loop closure detection. Finally, a factor graph optimization strategy is employed for global pose optimization, further improving localization accuracy. Simulation analysis proves that the trajectory generated by LIS-VIS-Fusion achieves high accuracy. Compared to LVI-SAM, the maximum reduction in trajectory error reaches 43.8%. The localization accuracy and mapping performance of this system are tested in four environments: a daytime long corridor, a nighttime long corridor, an evening outdoor setting, and an excavator construction site. Experiments demonstrate that LIS-VIS-Fusion exhibits superior localization accuracy and mapping quality across various extreme environments, with the system boasting high stability and strong environmental adaptability. This provides reliable technical support for the precise operation of unmanned intelligent excavator.

Förderung

  • Fundamental Research Funds for the Central Universities
  • Special Funds of the Tongji University for “Sino-German Cooperation 2.0 Strategy”

Zitieren

APA

Wang, Z., & Liu, G. (2026). Research on excavator environment perception and localization system based on multi-sensor data fusion. Measurement Science and Technology, 37(22), 226304. https://doi.org/10.1088/1361-6501/ae65cc

Vancouver

Wang Z, Liu G. Research on excavator environment perception and localization system based on multi-sensor data fusion. Measurement Science and Technology. 2026;37(22):226304. doi:10.1088/1361-6501/ae65cc

BibTeX
@article{Wang2026,
  author = {Wang, Ziwei and Liu, Guangjun},
  title = {Research on excavator environment perception and localization system based on multi-sensor data fusion},
  journal = {Measurement Science and Technology},
  year = {2026},
  volume = {37},
  number = {22},
  pages = {226304},
  doi = {10.1088/1361-6501/ae65cc}
}
Kostenlos starten →