A multi-sensor fusion approach for maritime autonomous surface ships berthing navigation perception

Port environment perceiving is a crucial step in ship autonomous berthing process. Multi-sensor fusion perception is particularly well-suited for scenarios involving ship offshore navigation. Recent advancements in high-precision sensing equipment have accelerated the development of autonomous navigation and remote control technologies. However, these advancements have also introduced challenges related to the complexity of data processing and high computational demands. Currently, there are limited studies focusing on lightweight perception systems specifically designed for the berthing tasks of maritime autonomous surface ships (MASS). This paper proposes a multi-sensor fusion approach for the perception of MASS during berthing navigation, based on a simultaneous localization and mapping (SLAM) framework, aimed at improving positioning accuracy and back-end processing speed. The method integrates real time kinematic (RTK), lightlaser detection and ranging (LiDAR), and inertial measurement unit (IMU) to mitigate the issue of front-end odometry drift, which is common in traditional SLAM applications for MASS. The back-end processing speed is enhanced through the optimization of an adaptive pose similarity evaluation method. Additionally, to balance perception accuracy and the computational load on shipborne equipment, a scanning context method is applied to reduce computational redundancy in back-end processing. Experimental results demonstrate that the proposed method outperforms traditional LiDAR SLAM. Specifically, the absolute pose error (APE) is improved by approximately 32.8%, with an average error of 0.304 m. The root mean square error (RMSE) is improved by 31.7%, with an average error of 0.339 m. The maximum and minimum errors recorded were respectively 0.535m and 0.087m. The average processing time per-frame is 98.1ms, indicating that the perception accuracy and real-time performance of the proposed method meet the requirements of a MASS berthing assistance system.

• Record URL:
  • https://doi.org/10.1016/j.oceaneng.2024.119965
• Record URL:
  • http://www.sciencedirect.com/science/article/pii/S0029801824033031
• Availability:
  • Find a library where document is available. Order URL: http://worldcat.org/issn/00298018
• Supplemental Notes:
  • © 2024 Elsevier Ltd. All rights are reserved, including those for text and data mining, AI training, and similar technologies. Abstract reprinted with permission of Elsevier.
• Authors:
  • Lu, Hongrui
  • Zhang, Yingjun
  • 0009-0001-1561-7682
  • Zhang, Chuang
  • Niu, Yihan
  • 0009-0002-3076-4469
  • Wang, Zhuolin
  • Zhang, Haoze
• Publication Date: 2025-1-15

Language

• English

Media Info

• Media Type: Web
• Features: References;
• Pagination: 119965
• Serial:
  • Ocean Engineering
  • Volume: 316
  • Issue Number: 0
  • Publisher: Pergamon
  • ISSN: 0029-8018
  • EISSN: 1873-5258
  • Serial URL: http://www.sciencedirect.com/science/journal/00298018

Subject/Index Terms

• TRT Terms: Autonomous vehicles; Data fusion; Docking; Navigation processes; Port operations; Ships
• Subject Areas: Data and Information Technology; Marine Transportation; Terminals and Facilities; Vehicles and Equipment;

Filing Info

• Accession Number: 01940886
• Record Type: Publication
• Files: TRIS
• Created Date: Dec 24 2024 4:45PM