水下SLAM技术:挑战、进展与未来方向
DOI:
CSTR:
作者:
作者单位:

1.中国船舶及海洋工程设计研究院;2.黑龙江职业学院

作者简介:

通讯作者:

中图分类号:

基金项目:


Underwater SLAM Technology: Challenges, Progress, and Future Directions
Author:
Affiliation:

1.Marine Design and Research Institute of China;2.Heilongjiang Polytechnic

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    水下同步定位与地图构建(SLAM)技术是水下机器人探索未知区域的重要手段,在复杂海洋环境中广泛应用。水下SLAM主要分为三种主流技术:前视声呐SLAM、水下光学SLAM及基于测深信息的SLAM。首先,对这些技术在海底地形测绘和深海资源勘探中的应用进行了总结,不同的技术显示出不同的适用性和准确性,其中多波束声呐技术因其高分辨率和大范围覆盖能力,在大规模海底地图构建中展现了优越性。然后,总结了人工智能技术等关键技术在处理复杂海底数据中的应用,例如通过神经网络优化的数据处理提高了地图构建的效率和精度。最后,对未来的SLAM技术进行了展望,未来通过算法优化和硬件发展,进一步提升数据处理速度和降低能耗。

    Abstract:

    Underwater Simultaneous Localization and Mapping (SLAM) technology is a crucial tool for underwater robots to explore unknown areas, widely applied in complex marine environments. Underwater SLAM is primarily divided into three mainstream technologies: forward-looking sonar SLAM, underwater optical SLAM, and depth information-based SLAM. Firstly, a summary of these technologies in the application of seabed topography surveying and deep-sea resource exploration is provided, showing varying suitability and accuracy among the technologies, with multibeam sonar technology demonstrating superiority in large-scale seabed mapping due to its high resolution and extensive coverage capabilities. Additionally, the application of key technologies such as artificial intelligence in processing complex underwater data is summarized, for instance, enhanced map construction efficiency and precision through neural network-optimized data processing. Lastly, the future of SLAM technology is envisioned, with further enhancements in data processing speed and energy consumption reduction through algorithm optimization and hardware development.

    参考文献
    相似文献
    引证文献
引用本文
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2024-04-29
  • 最后修改日期:2024-05-16
  • 录用日期:2024-06-17
  • 在线发布日期:
  • 出版日期:
文章二维码