当前位置>主页 > 期刊在线 > 计算机技术 >

计算机技术23年7期

基于LSTM神经网络的海缆保护区船舶搁浅预测模型
吴仕良 1 ,林业茂 1 ,张雪峰 1 ,黄拨 2 ,刘莉莉 1
(1. 中国人民解放军 75841 部队,海南 海口 570100;2. 三亚中科遥感研究所,海南 三亚 572022)

摘  要:针对我国沿海地区,船舶搁浅可能损坏海底电缆的问题。设计一种基于雷达网和 AIS 数据的轨迹预测模型。该模型通过采集船舶的经纬度,航速航向等信息,预测五分钟内的船舶位置。使用 LSTM 神经网络对船舶五分钟内的轨迹进行预测,准确预测船舶轨迹可以及时发出预警信息,降低船舶触缆的风险。5 分钟内船舶轨迹点平均距离误差约为 124 米,5 分钟时刻,船舶坐标点平均距离误差约为 185 米。


关键词:船只搁浅;轨迹预测;LSTM



DOI:10.19850/j.cnki.2096-4706.2023.07.021


中图分类号:TP391                                       文献标识码:A                                   文章编号:2096-4706(2023)07-0082-05


Prediction Model of Ship Grounding in Submarine Cable Protection Area Based on LSTM Neural Network

WU Shiliang1, LIN Yemao1, ZHANG Xuefeng1, HUANG Bo2, LIU Lili 1

(1.75841 Unit of the PLA, Haikou 570100, China; 2.Sanya Zhongke Remote Sensing Research Institute, Sanya 572022, China)

Abstract: In view of the problem of coastal areas in our country, the ship grounding may damage the submarine cables. This paper designs a trajectory prediction model based on radar net and AIS data. The model can predict the ship's position in real time within five minutes by collecting information such as the ship's latitude and longitude, speed and course. It uses the LSTM neural network to predict the trajectory of the ship within five minutes, and the accurate prediction of the ship trajectory can send out early warning information in time and reduce the risk of the ship touching the cables. The average distance error of ship trajectory point within 5 minutes is about 124 meters, and the average distance error of the ship coordinate point at the time of 5 minutes is about 185 meters.

Keywords: ship grounding; trajectory prediction; LSTM 


参考文献:

[1] PALMA A T,BOGORNY V,KUIJPRS B,et al. A clustering-based approach for discovering interesting places in trajectories [C]//SAC '08:Proceedings of the 2008 ACM symposium on Applied computing.New York:ACM,2008:863-868.

[2] VESPE M,VISENTINI I,BRYAN K,et al. Unsupervised learning of maritime traffic patterns for anomaly detection [C]//9th IET Data Fusion & Target Tracking Conference (DF&TT 2012):Algorithms & Applications. London:IET,2012:1-5.

[3] DE VRIES G K D,VAN SOMEREN M. Machine learning for vessel trajectories using compression,alignments and domain knowledge [J].Expert Systems with Applications,2012,39(18):13426-13439.

[4] DOUGLAS D H,PEUCKER T K. Algorithms for the reduction of the number of points required to represent a digitized line or its caricature [J].Cartographica:The International Journal for Geographic Information and Geovisualization,2003,10(2):112-122.

[5] 刘兴龙,初秀民,马枫,等 .AIS 报文异常动态信息甄别方法 [J]. 交通运输工程学报,2016,16(5):142-150.

[6] BIERE A,FROHLICH A. Evaluating CDCI. Variable Scoring Schemes [C]//Proceedings of the 18th International Conference on Theory and Applications of Satisfiability Testing.Bioinformatics: Springer International Publishing,2015:405-422.

[7] CHEN J. A Dynamic Phase Selection Strategy for Satisfiability Solvers [J].Computer Science,2012,13(5):208-213.

[8] PERERA L P,OLIVEIRA P,SOARES C G. Maritime Traffic Monitoring Based on Vessel Detection,Tracking,State Estimation,and Trajectory Prediction [J].IEEE Transactions on Intelligent Transportation Systems,2012,13(3):1188-1200.

[9] 权波,杨博辰,胡可奇,等 . 基于 LSTM 的船舶航迹预测模型 [J]. 计算机科学,2018,45(S2):126-131.

[10] 赵万香 . 基于马尔科夫预测法的车道偏离预警系统研究[D]. 西安:长安大学,2014.

[11] 孙平阔 . 海底光缆路由设计影响因素 [J]. 电信工程技术与标准化,2022,35(2):76-80.

[12] 叶胤,王超,莫仁芸.海底光缆通信系统技术发展分析 [J].广东通信技术,2021,41(1):19-23.

[13] 魏巍,王增,张伟,等 . 新一代海底光缆综合信息传输网技术发展现状 [J]. 光通信技术,2022,46(4):61-67.

[14] 刘兴华 . 海底通信光缆水下设备原理及设计研究 [J]. 广东通信技术,2022,42(5):68-72.

[15] 卢聃 . 海底光缆突围在即 [J]. 产城,2021(11):70-71.

[16] 陈宇俊,解江,张泽,等 . 海底光缆环境影响因素综述 [J].电子产品可靠性与环境试验,2018,36(S1):232-236.

[17] 施炎,李文川,丛丕宇 . 海底光缆风险警告解决方案 [J].电信技术,2018(11):54-58.


作者简介:吴仕良(1988—),男,汉族,广西平南人,助理工程师,本科,主要研究方向 : 光纤通信;林业茂(1986—),男,汉族,海南万宁人,助理工程师,硕士研究生,主要研究方向:光纤通信;张雪峰(1983—),男,汉族,湖南衡阳人,工程师,硕士研究生,主要研究方向:通信工程;黄拨(1985—),男,汉族,湖南长沙人,助理研究员,硕士研究生,主要研究方向 : 遥感技术;刘莉莉(1988—),女,汉族,福建泉州人,本科,助理工程师,主要研究方向:光纤通信。