基于语义的船舶行为动态推理机制  

Mechanism of Ship Behavior Dynamic Reasoning Based on Semantics

作　　者：文元桥[1,3,4] 张义萌 黄亮[1,3] 周春辉[1,2,3] 肖长诗[2,4] 张帆[2,4] WEN Yuanqiao;ZHANG Yimeng;HUANG Liang;ZHOU Chunhui;XIAO Changshi;ZHANG Fan(National Engineering Research Center for Water Transport Safety, Wuhan University of Technology, Wuhan 430063, China;School of Navigation, Wuhan University of Technology, Wuhan 430063, China;Intelligent Transportation Systems Research Center, Wuhan University of Technology, Wuhan 430063, China;Hubei Key Laboratory of Inland Shipping Technology, Wuhan University of Technology, Wuhan 430063, China)

机构地区：[1]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 [2]武汉理工大学航运学院,湖北武汉430063 [3]武汉理工大学智能交通系统研究中心,湖北武汉430063 [4]武汉理工大学内河航运技术湖北省重点实验室,湖北武汉430063

出　　处：《中国航海》2019年第3期34-39,50,共7页Navigation of China

基　　金：国家自然科学基金(51679180;51579204;51709218);武汉大学测绘遥感信息工程国家重点实验室开放基金(17I03);国家重点研发计划(2018YFC1407405)

摘　　要：为实现对船舶行为的深入挖掘,建立船舶行为模型,根据模型需求引入语义模型和动态贝叶斯网络,形成基于语义的船舶行为动态推理机制。使用语义网络实现复杂态势下船舶行为领域知识的形式化描述与共享;将语义网络结构转换为动态贝叶斯网络结构并采用水上交通大数据进行参数学习;使用动态贝叶斯网络推理不确定性信息,挖掘深层次的船舶行为和事件。基于长江渡船靠离泊行为的实例验证表明:该船舶行为动态推理模型能准确地识别并预测船舶的动态行为,实现船舶行为的辨识与预警。The ship behavior model is established for deep mining of ship behavior. According to the model requirements, the semantic model and the dynamic bayesian network are built to form a semantic-based ship behavior dynamic reasoning mechanism. The semantic network is used to describe and share knowledge of ship behavior. The semantic network structure is transformed into a dynamic Bayesian network structure. The network parameter learning is performed with water traffic big data. The dynamic Bayesian network is used to perform reasoning of the information of uncertainty and find the hidden ship behavior. The arrival/departure behaviors of the ferries in the Yangtze River are studied with the model, which proves that the model is good for ship identification and early warning of the improper ship behavior.

关 键 词：水路运输 行为动态推理 动态贝叶斯网络 船舶 语义网络 船舶行为 

分 类 号：U692[交通运输工程—港口、海岸及近海工程]