由国际海事组织（IMO）、国际电信联盟（ITO）、国际电工委员会（IEC）等组织牵头研发的以现代通信技术、电子信息技术等为核心的船舶自动识别系统（Automatic Identification System, AIS）几乎安装在所有在籍的船舶上，AIS数据在船舶的定位、动态轨迹上有着重要的作用。针对现有用于AIS数据轨迹插值的方法存在平滑性能较差、精度无法统一等缺点，基于GRNN神经网络的非线性映射能力及学习速度极高的优势，提出一种用于AIS数据插值的L-GRNN算法。仿真研究和实船数据验证表明，该方法的精度、平滑性能较好，与传统的AIS数据插值方法相比，同时具备精度、平滑性能，具有良好的工程适用价值。
The automatic identification system (AIS) is developed by the International Maritime Organization (IMO), the International Telecommunication Union (ITO), the International Electrotechnical Commission (IEC) and other organizations, AIS has been installed on almost all registered ships. AIS data plays an important role in the positioning and dynamic trajectory of ships. In view of the shortcomings of the existing methods for AIS data trajectory interpolation, such as poor smoothing performance and inconsistent accuracy, a new method l-grnn algorithm for AIS data interpolation is proposed based on the nonlinear mapping ability and high learning speed of GRNN neural network. Through simulation research and verification of real ship data, the accuracy and smoothing performance of this method are better. Compared with the traditional AIS data interpolation method, the method has accuracy and smooth performance, and has good engineering application value.
2021,43(9): 117-121 收稿日期：2020-09-29
DOI：10.3404/j.issn.1672-7649.2021.09.023
分类号：U675.79
基金项目：工信部智能船舶测试与验证项目([2018] 473)
作者简介：蒲实(1995-)，男，硕士研究生，研究方向为智能船舶测试与分析技术
参考文献：
[1] ZHANG Li-ye, MENG Qiang, XIAO Zhe, et al. A novel ship trajectory reconstruction approach using AIS data[J]. Ocean Engineering, 2018: 159
[2] 交通大辞典编辑委员会. 交通大辞典(精)[M]. 上海：上海交通大学出版社, 2005.
[3] 周鹏, 陈金海, 林乾锋, 等. 基于AIS的轨迹插值方法[J]. 集美大学学报(自然科学版), 2018, 23(06): 443–447
[4] 薛晗, 邵哲平, 潘家财,等. 基于文化萤火虫算法-广义回归神经网络的船舶交通流量预测[J]. 上海交通大学学报, 2020, 54(04): 421–429
[5] Imagination Technologies Limited; Patent Issued for Interpolating A Sample Value In A Data Array By Linear Interpolation Of Surrounding Interpolated Values (USPTO 10, 719, 576)[J]. Technology News Focus, 2020.
[6] 李艳. 径向基函数及其应用[D]. 大连：大连理工大学，2009.
