基于BLSTM-RNN的船舶轨迹修复方法

doi:10.3969/j.issn.1674-0696.2019.10.02

重庆交通大学学报（自然科学版） ›› 2019, Vol. 38 ›› Issue (10): 7-12.DOI: 10.3969/j.issn.1674-0696.2019.10.02

• 交通+大数据人工智能 • 上一篇下一篇

基于BLSTM-RNN的船舶轨迹修复方法

王贵槐1，钟诚2，初秀民2，张代勇2

(1. 武汉交通职业学院，湖北武汉 430065； 2. 武汉理工大学国家水运安全工程技术研究中心，湖北武汉 430063)

收稿日期:2018-05-19 修回日期:2018-09-14 出版日期:2019-10-14 发布日期:2019-10-14
作者简介:王贵槐(1960—)，男，湖北武汉人，副教授，博士，主要从事交通信息化方面的研究。E-mail：734832417@qq.com。通信作者：钟诚（1987—），男，湖北武汉人，博士研究生，主要从事水上智能交通研究。E-mail：zcplace@whut.edu.cn。
基金资助:
武汉市科技计划项目(2017010202010132); 武汉理工大学研究生自主创新基金项目(2017-YB-021)

Ship Trajectory Restoration Method Based on BLSTM-RNN

WANG Guihuai1, ZHONG Cheng2, CHU Xiumin2, ZHANG Daiyong2

(1. Wuhan Technical College of Communications, Wuhan 430065, Hubei, P. R. China; 2. National Engineering Research Center for Water Transport Safety, Wuhan University of Technology, Wuhan 430063, Hubei, P. R. China)

Received:2018-05-19 Revised:2018-09-14 Online:2019-10-14 Published:2019-10-14

摘要/Abstract

摘要： 针对内河干线船舶AIS轨迹数缺失问题，提出一种基于双向长短时记忆网络(BLSTM-RNN)模型的船舶轨迹数据修复方法。通过利用船舶轨迹上下文信息及其他回传特征作为模型输入，构建两层的双向循环神经网络(RNN)模型。在模型输入上，采用相关性分析及序列自相关系数，确定船舶轨迹点相关变量及轨迹序列自相关滞后值；在模型结构上，以ACC率为指标对模型超参数值进行合理设置，以长江干线航道武汉段及重庆段船舶轨迹数据为样本，对模型进行实证验证。实验结果表明：与线性及其他机器学习方法相比BLSTM-RNN方法在精度上有一定提升；在武汉段顺直河段实验中，将修复误差控制在15 m量级内，远低于其他非线性方法的50 m量级；在重庆复杂河段内，可将修复误差控制在10 m量级；模型解决了传统方法在长距离丢失点上精度缺失的问题，在20个连续点丢失的情况上，将修复误差降低至50m量级。

关键词: 船舶工程, 双向长短时记忆网络(BLSTM), 循环神经网络(RNN), 船舶轨迹修复, 船舶自动驾驶

Abstract: Aiming at the problem of missing AIS trajectory number of inland waterway trunk ships, a method of ship trajectory data repair method based on bidirectional long-term and short-term memory network (BLSTM-RNN) model was proposed. A two-layer bidirectional recurrent neural network (RNN) model was constructed by using ship trajectory context information and other return features as model input. In the model input, correlation analysis and sequence autocorrelation coefficient were used to determine the correlation variables of ship trajectory points and the autocorrelation lag value of ship trajectory sequence. In the model structure, the super-parametric values of the model were reasonably set with the ACC rate as the index. Taking the ship trajectory data of Wuhan section and Chongqing section of the Yangtze River trunk waterway as samples, empirical verification of the proposed model was carried out. The experimental results show that, compared with linear and other machine learning methods, the accuracy of BLSTM-RNN method is improved to a certain extent. In the experiment of straight-line reach in Wuhan section, the repair error is controlled within 15 m magnitude, which is much lower than that of other non-linear methods. In the complex reach of Chongqing section, the repair error can be controlled in the order of 10m. In addition, the proposed model solves the problem of accuracy loss of traditional methods at long-distance loss points, and reduces the repair error to 50 m magnitude when 20 continuous points are lost.

Key words: marine engineering, bidirectional long-term and short-term memory network (BLSTM), recurrent neural network (RNN), ship trajectory restoration, ship autopilot

中图分类号:

U675.7

王贵槐1，钟诚2，初秀民2，张代勇2. 基于BLSTM-RNN的船舶轨迹修复方法[J]. 重庆交通大学学报（自然科学版）, 2019, 38(10): 7-12.

WANG Guihuai1, ZHONG Cheng2, CHU Xiumin2, ZHANG Daiyong2. Ship Trajectory Restoration Method Based on BLSTM-RNN[J]. Journal of Chongqing Jiaotong University(Natural Science), 2019, 38(10): 7-12.

参考文献

［1］初秀民, 刘潼, 马枫, 等. 山区航道 AIS 信号场强分布特性［J］. 交通运输工程学报, 2014(6):117-126.
CHU Xiumin, LIU Tong, MA Feng, et al. Distribution characteristic of AIS signal field intensity along mountainous waterway［J］. Journal of Traffic &
Transportation Engineering, 2014, 6(1):117-126.
［2］ OH S R, SUN J. Path following of under actuated marine surface vessels using line-of-sight based model predictive control［J］. Ocean Engineering, 2010,37
(2): 289-295.
［3］马建文,涂兴华,吴小峰,等. 多自由度本船操纵运动仿真［J］. 上海海事大学学报, 2016,37(4): 32-35.
MA Jianwen, TU Xinghua, WU Xiaofeng, et al. Simulation on ownship manoeuvring motion with multi-degrees of freedom［J］. Journal of Shanghai Maritime University,
2016, 37(4):32-35.
［4］ HU Q , CAI F , YANG C , et al. An algorithm for interpolating ship motion vectors［J］. Transnav the International Journal on Marine Navigation & Safety of
Sea Transportation, 2014, 8(1):35-40.
［5］徐婷婷,柳晓鸣,杨鑫.基于BP神经网络的船舶航迹实时预测［J］.大连海事大学学报,2012,38(1):9-11.
XU Tingting, LIU Xiaoming, YANG Xin. BP neural network-based ship track real-time prediction ［J］. Journal of Dalian Maritime University. 2012,38(1):9-11.
［6］王体迎, 时鹏超, 刘蒋琼, 等. 基于门限递归单元循环神经网络的交通流预测方法研究［J］. 重庆交通大学学报(自然科学版), 2018, 37(11):76-82.
WANG Tiying, SHI Pengchao, LIU Jiangqiong, et al. Research on traffic flow prediction method based on gated recurrent unit recurrent neural network ［J］. Journal
of Chongqing Jiaotong University (Natural Science). 2018, 37(11):76-82.
［7］ CHUNYAN L, JIANCHUN L , LINYUAN K . Performance comparison between GA-BP neural network and BP neural network［J］. Chinese Journal of Health Statistics,
2013, 30(2):173-494.
［8］ KANARACHOS S, CHRISTOPOULOS S , CHRONEOS A , et al. Detecting anomalies in time series data via a deep learning algorithm combining wavelets, neural networks
and Hilbert transform［J］. Expert Systems with Applications, 2017, 85, 292-304.
［9］ REINHART R F, STEIL J J. A constrained regularization approach for input-driven recurrent neural networks［J］. Differential Equations & Dynamical Systems,
2011, 19(1-2):27-46.
［10］ BASHEER I A. Artificial neural networks: fundamentals, computing, design, and application［J］. Journal of Microbiological Methods, 2000, 43(1):3-31.
［11］ ZHANG X , GU N , YASRAB R , et al. GT-SGD: A novel gradient synchronization algorithm in training distributed recurrent neural network language models
［C］//International Conference on Networking & Network Applications. Kathmanda:IEEE Computer Society, 2017. ［1］初秀民, 刘潼, 马枫, 等. 山区航道 AIS 信号场强分布特性［J］. 交通运输工程学报, 2014(6):117-126.
CHU Xiumin, LIU Tong, MA Feng, et al. Distribution characteristic of AIS signal field intensity along mountainous waterway［J］. Journal of Traffic &
Transportation Engineering, 2014, 6(1):117-126.
［2］ OH S R, SUN J. Path following of under actuated marine surface vessels using line-of-sight based model predictive control［J］. Ocean Engineering, 2010,37
(2): 289-295.
［3］马建文,涂兴华,吴小峰,等. 多自由度本船操纵运动仿真［J］. 上海海事大学学报, 2016,37(4): 32-35.
MA Jianwen, TU Xinghua, WU Xiaofeng, et al. Simulation on ownship manoeuvring motion with multi-degrees of freedom［J］. Journal of Shanghai Maritime University,
2016, 37(4):32-35.
［4］ HU Q , CAI F , YANG C , et al. An algorithm for interpolating ship motion vectors［J］. Transnav the International Journal on Marine Navigation & Safety of
Sea Transportation, 2014, 8(1):35-40.
［5］徐婷婷,柳晓鸣,杨鑫.基于BP神经网络的船舶航迹实时预测［J］.大连海事大学学报,2012,38(1):9-11.
XU Tingting, LIU Xiaoming, YANG Xin. BP neural network-based ship track real-time prediction ［J］. Journal of Dalian Maritime University. 2012,38(1):9-11.
［6］王体迎, 时鹏超, 刘蒋琼, 等. 基于门限递归单元循环神经网络的交通流预测方法研究［J］. 重庆交通大学学报(自然科学版), 2018, 37(11):76-82.
WANG Tiying, SHI Pengchao, LIU Jiangqiong, et al. Research on traffic flow prediction method based on gated recurrent unit recurrent neural network ［J］. Journal
of Chongqing Jiaotong University (Natural Science). 2018, 37(11):76-82.
［7］ CHUNYAN L, JIANCHUN L , LINYUAN K . Performance comparison between GA-BP neural network and BP neural network［J］. Chinese Journal of Health Statistics,
2013, 30(2):173-494.
［8］ KANARACHOS S, CHRISTOPOULOS S , CHRONEOS A , et al. Detecting anomalies in time series data via a deep learning algorithm combining wavelets, neural networks
and Hilbert transform［J］. Expert Systems with Applications, 2017, 85, 292-304.
［9］ REINHART R F, STEIL J J. A constrained regularization approach for input-driven recurrent neural networks［J］. Differential Equations & Dynamical Systems,
2011, 19(1-2):27-46.
［10］ BASHEER I A. Artificial neural networks: fundamentals, computing, design, and application［J］. Journal of Microbiological Methods, 2000, 43(1):3-31.
［11］ ZHANG X , GU N , YASRAB R , et al. GT-SGD: A novel gradient synchronization algorithm in training distributed recurrent neural network language models
［C］//International Conference on Networking & Network Applications. Kathmanda:IEEE Computer Society, 2017.

[1]	李胜永1，张智华1，王胜男2，王孟2. 面向交通工具金属材料的缺损识别算法[J]. 重庆交通大学学报（自然科学版）, 2021, 40(07): 136-144.
[2]	张爱锋，刘少康，姚苗苗，甄春博. 船桥碰撞结构损伤及船撞力影响因素分析[J]. 重庆交通大学学报（自然科学版）, 2021, 40(03): 121-127.
[3]	赵月林，门茂峰，孙壮. 基于分段式双幂次趋近律的迭代滑模航向控制[J]. 重庆交通大学学报（自然科学版）, 2021, 40(02): 136-142.
[4]	赵月林，马征，孙壮. 考虑舵机时滞的船舶航向控制系统研究[J]. 重庆交通大学学报（自然科学版）, 2020, 39(12): 124-129.
[5]	杨敬东1，黎扬武1，刘美山2,黄兵3. 60 m3铰接式对开泥驳铰链系统受力分析及优化设计[J]. 重庆交通大学学报（自然科学版）, 2020, 39(12): 130-134.
[6]	张杰1，2，邓羽1,2，彭中波1，石利勇1. 基于声子晶体的船机减振特性研究[J]. 重庆交通大学学报（自然科学版）, 2020, 39(09): 140-145.
[7]	郝刚1，2，金涛1. 基于CAN总线的船用动力锂电池控制系统设计[J]. 重庆交通大学学报（自然科学版）, 2020, 39(04): 123-128.
[8]	谢新连，何平，何傲，辛剑英. 复杂水域船舶避碰路径规划研究[J]. 重庆交通大学学报（自然科学版）, 2019, 38(12): 1-7.
[9]	杨敬东，何瑞峰，刘文彬. 海巡163轮锚链绞车下支撑船体结构加强和强度计算[J]. 重庆交通大学学报（自然科学版）, 2019, 38(08): 111-116.
[10]	蒋效彬1，谭家万2，任鸿翔1，邱绍杨1. 基于虚拟现实技术的船用锚机交互仿真系统[J]. 重庆交通大学学报（自然科学版）, 2019, 38(06): 12-16.
[11]	邱绍杨1，谭家万2，任鸿翔1，蒋效彬1. 尾抛式救生艇两种释放过程的建模与分析[J]. 重庆交通大学学报（自然科学版）, 2019, 38(06): 115-121.
[12]	廉静静1，杨晓1，尹勇2. 基于数值分析的船舶斜航运动水动力研究[J]. 重庆交通大学学报（自然科学版）, 2018, 37(07): 113-117.
[13]	杨敬东，刘永臻，雷林. 基于SHIPFLOW内河船阻力预报研究[J]. 重庆交通大学学报（自然科学版）, 2018, 37(07): 118-122.
[14]	张婧，邵龙，吴洵，闫岩. 张紧式系泊深水FPSO的动力特性及系泊布置方式研究[J]. 重庆交通大学学报（自然科学版）, 2018, 37(04): 121-127.
[15]	张继祥，蒋飞，杨泮，唐博. 船用钢Q345D中温塑性成形本构方程建立及分析[J]. 重庆交通大学学报（自然科学版）, 2018, 37(03): 121-126.

基于BLSTM-RNN的船舶轨迹修复方法

Ship Trajectory Restoration Method Based on BLSTM-RNN

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics