为实现船舶航行线路全面监测，自动实现偏航修正，保证船舶安全行驶，设计人工智能的船舶航线自动监测系统。通过海图集成模块采集船舶航行相关数据信息，并依据信息生成电子航海图数据。船载通信模块通过网络基站、收发站、船岸通信链路等网络传输纽带，将航行相关数据信息以及航海图数据传输到航线设计模块；航线设计模块依据航行数据，设计航行前的船舶航线以及航行过程中的航线；航线监测模块采用基于长短期记忆网络和注意力机制网络模型，预测船舶航行线路，判断航行线路与航行设计模块设定航线是否一致，如果不一致，采用航线自动解算方法修正偏航段，并可视化呈现航行时间、航向速度等监测结果。测试结果显示：该系统能够依据航行目的地，设计船舶航行线路，并生成电子海图；船舶航线预测误差结果均低于（0°0.5`E，0°0.35`N），且可对偏航航段进行偏航提醒，同时自动修正偏航线路，实现船舶航线的全面监测。
In order to realize the comprehensive monitoring of shipping routes, automatically realize yaw correction, and ensure the safe sailing of ships, an artificial intelligence automatic monitoring system for shipping routes is designed. Collect navigation-related data information through chart integration module, and generate electronic chart data according to the information. The shipboard communication module transmits navigation-related data information and chart data to the route design module through network transmission links such as network base station, transceiver station and ship-shore communication link. The route design module designs the route before the voyage and the route during the voyage according to the voyage data. Route monitoring module based on short - and long-term memory and attention mechanism network model, forecast of navigation line, determine shipping lines set routes are consistent with the navigation design module, if inconsistent, route automatically calculating method modified partial segment, and visualization sailing time, heading speed monitoring results, etc. The test results show that the system can design the navigation route and generate electronic chart according to the navigation destination. The error results of ship route prediction are all lower than (0° 0.5’e, 0° 0.35’n). The yaw section is reminded, and the yaw line is automatically corrected to comprehensively monitor the ship route.
2022,44(21): 148-151 收稿日期：2022-07-19
DOI：10.3404/j.issn.1672-7649.2022.21.030
分类号：TP399
基金项目：长沙市科技局2020年度指导性科技计划项目（kzd2011011）
作者简介：夏峰(1978-),男,硕士,研究方向为人工智能与大数据
参考文献：
[1] 周毅, 何金平, 华先亮, 等. 智能船舶船岸一体视频远程管理系统设计应用[J]. 船海工程, 2021, 50(3): 113–116
ZHOU Yi, HE Jinping, HUA Xianliang, et al. Design and application of a ship-shore integrated video management system[J]. Ship & Ocean Engineering, 2021, 50(3): 113–116
[2] 胡琦, 朱国情, 陈于涛. 船舶监控系统运行数据抽取与分析方案设计[J]. 船海工程, 2020, 49(3): 112-116+120.
[3] 常婧, 柳晓鸣, 李梦蕊. 基于SVR的船舶航行安全评估模型[J]. 上海海事大学学报, 2020, 41(2): 21–26
CHANG Jing, LIU Xiaoming, LI Mengrui. Ship navigation safety assessment model based on SVR[J]. Journal of Shanghai Maritime University, 2020, 41(2): 21–26
[4] 张阳, 高曙, 何伟, 等. 基于多模型融合的内河船舶航行轨迹预测方法[J]. 中国机械工程, 2022, 33(10): 1142–1152
[5] 祁林, 渠俊锋, 司文杰, 等. 基于RBF神经网络的水面船舶轨迹跟踪控制[J]. 船舶工程, 2021, 43(1): 95–101+118
[6] 李明峰, 王胜正, 谢宗轩. 恶劣气象海况下船舶航线的多变量多目标优化建模[J]. 中国航海, 2020, 43(2): 14–19+30
[7] 潘伟, 董其林, 许小卫, 等. 考虑通航环境因素的智能船舶航线规划[J]. 上海海事大学学报, 2021, 42(3): 76–84
[8] 韩志豪, 汪益兵, 张宇, 等. 基于深度强化学习的船舶航线自动规划[J]. 中国航海, 2021, 44(1): 100–105
HAN Zhihao, WANG Yibing, ZHANG Yu, et al. Automatic ship route planning based on deep reinforcement learning[J]. Navigation of China, 2021, 44(1): 100–105
