作者单位:

1. 四川师范大学 计算机科学学院，四川 成都 610066;2. 四川城市职业学院信息技术学院，四川 成都 6101011. School of computer science and technology, Sichuan Normal University, Chengdu 610066, China;2. School of information technology, Urban Vocational College of Sichuan, Chengdu 610101, China

大数据分析;航迹拟合;数据采集;插值法;中值滤波;曲线选择big data analysis; track fitting; data acquisition; interpolation method; median filtering; curve selection

船舶航行过程中所产生的海量航迹数据可通过拟合方法完成压缩，由此降低航迹数据存储的压力，考虑航迹数据采集过程中的误差与数据缺失问题，设计了大数据分析的船舶航迹拟合方法。以船舶自动识别系统为基础，根据参考物标采集船舶航迹数据。考虑所采集船舶航迹数据在时空2个维度上的误差与数据缺失问题，采用大数据分析技术中的三次样条插值法与中值滤波法，对所采集的船舶航迹数据实施预处理；针对预处理后的船舶航迹数据，通过坐豹换、曲线选择以及拟合计算等过程完成船舶航迹拟合处理。实验结果显示，所研究方法能够在不影响航迹数据局部趋势的条件下有效抑制航迹数据的波动性，得到更高精度的航迹拟合结果。The massive track data generated during ship navigation can be compressed by fitting method, so as to reduce the pressure of track data storage. Considering the error and data loss in the process of track data acquisition, the ship track fitting method of big data analysis is studied. Based on the ship automatic identification system, the ship track data is collected according to the reference object. Considering the error and data missing of the collected ship track data in two dimensions of time and space, the cubic spline interpolation method and median filter method in big data analysis technology are used to preprocess the collected ship track data. For the preprocessed ship track data, the ship track fitting processing is completed through the processes of coordinate conversion, curve selection and fitting calculation. The experimental results show that the proposed method can effectively suppress the volatility of track data without affecting the local trend in track data, and obtain higher precision track fitting results.

参考文献：

[1] 鄢博冉, 高大为, 朱永生, 等. 基于低信噪比AIS数据的船舶航迹精准预测[J]. 船舶工程, 2021, 43(10): 111–117

[2] 郭杰, 刘轶华, 马利华. 基于多模态快速非奇异终端滑模的船舶航迹纲自抗扰控制[J]. 中国航海, 2020, 43(2): 7–13

[3] 陈天元, 袁伟, 俞孟蕻. 基于显式模型预测控制的无人船航迹控制方法[J]. 船舶工程, 2020, 42(9): 122–127

[4] 张杰, 王刚, 姚小强, 等. 双向RNN下的航迹拟合模型研究[J]. 计算机科学, 2019, 46(S2): 58–61+88

[5] 陈影玉, 索永峰, 杨神化. 基于灰狼优化支持向量回归的船舶航迹预测[J]. 上海海事大学学报, 2021, 42(4): 20–25+46

[6] 夏辉宇, 刘安林, 程佳. 基于AIS大数据的南京长江大桥水域规律航迹研究[J]. 水运工程, 2021(1): 189–195