高级检索

声速剖面简化与模板插值结合的多波束测深快速声线跟踪方法

A Fast Sound Ray Tracing Method for Multibeam Bathymetry Combining Sound Velocity Profile Simplification and Template Interpolation

  • 摘要: 声线跟踪时高密度的声速剖面(sound velocity profile, SVP)会因逐层追加导致跟踪过程耗时较大,不利于海量多波束测深数据快速高精度计算。因此,本文提出一种SVP简化与模板插值结合的快速声线跟踪算法,首先以等效SVP作为约束,引入三角剖分思想,采用加权平均法灵活结合角度成本和距离成本,去除SVP冗余点,减少SVP层数,在保留关键声学特征的同时,确保声线跟踪的精度;然后,利用简化后的SVP构建虚拟波束声线跟踪模板,进而根据实际波束入射角及旅行时,采用反距离加权插值法计算波束的深度和水平距离,从而代替常规方法大量分层累加的计算过程,提升海量波束声线跟踪效率。结果表明,本方法与传统常梯度声线跟踪方法相比,在平均水深为32 m的浅水区域,最大水深偏差约为0.011 m,计算效率提升2.9倍;在平均水深为4100 m的深水区域,最大水深偏差约为0.40 m,计算效率提升16.7倍,在保证声线跟踪精度的同时大幅提升计算效率。研究方法可为无人船、自主水下航行器等平台的实时快速多波束数据处理提供参考。

     

    Abstract: When performing sound ray-trace process of the multibeam echo-sounder system, high-density sound velocity profiles (SVPs) usually cause significant time consumption in the tracing process due to layer-by-layer accumulation, which is unfavorable for fast and high-precision calculations of massive multibeam bathymetric data. Therefore, this paper proposes a fast sound ray-trace algorithm combining SVP simplification and template interpolation. First, using equivalent SVPs as constraints, the idea of triangulation is introduced, and a weighted average method is flexibly used to combine angular cost and distance cost to remove redundant points in the SVP and reduce the number of SVP layers, ensuring the accuracy of sound ray tracing while preserving key acoustic features. Then, the simplified SVP is used to construct a virtual beam sound ray-trace template. According to the actual beam incident angle and propagation time, the inverse distance weighted interpolation method is adopted to calculate the depth and horizontal distance of the beam to be tracked, replacing the conventional calculation process of massive layered accumulation and improving the efficiency of sound ray tracing for massive beams. The results show that compared with the traditional constant gradient sound ray tracing method, this method has a water depth deviation of approximately 0.34‰ in the shallow water area with an average water depth of 32 m, and the calculation efficiency is increased by 2.9 times. In the deep water area with an average water depth of 4100 m, the water depth deviation is approximately 0.1‰, and the calculation efficiency is increased by 16.7 times, significantly improving the calculation efficiency while ensuring the accuracy of sound ray tracing. The research method can provide a reference for real-time and fast multibeam data processing on platforms such as unmanned surface vessels and autonomous underwater vehicles.

     

/

返回文章
返回