摘要
根据不同情况下单锚泊船舶运动规律,提出相应的基于AIS信息的单锚泊船舶走锚监控预警方法.对于风、流较小时的船舶走锚监控,首先,采用一种基于相似度曲线且其Eps邻域阈值可调的改进密度空间聚类算法(DBSCAN),对原始锚泊船舶位置信息进行去噪处理;然后,对去噪后的数据进行圆曲线拟合,求出轨迹圆心并视为锚位点;最后,将其位置变化以及船速变化一并作为走锚的判定依据来判断船舶是否走锚并报警.对于风、流较大时的锚泊船走锚监控,则先确定锚泊船"∞"型运动的两个极限点的位置;然后,基于该两个极限点位置求出锚位点;最后,根据船位距锚位点的距离变化及船速变化进行走锚分析和判断.实例和模拟分析证明了所提方法的有效性.
According to the motion law of single anchored ship under different wind and current conditions, the corresponding monitoring and early warning method for single anchored ship was proposed based on AIS information. For ships with low wind and current, firstly, an improved density based spatial clustering of applications with noise method (DBSCAN) based on similarity curve with adjustable Eps neighborhood threshold was used to denoise the original mooring ship position information;then, the circular curve fitting of de-noised AIS data was conducted to find out the center of the trajectory as anchor point;finally, the changes of position and ship speed were taken as the judgment criterions of anchoring to judge whether the ship is anchoring and alarm. For ships with large wind and current, the position of the two limit points of the “infinity” motion of the anchored ship was determined firstly;then, anchor points were determined based on the above two limit points;finally, anchor analysis and judgement were made according to the distance change between ship position and anchor position, as well as the ship speed variation. Examples and simulation analysis prove the effectiveness of proposed method.
作者
肖进丽
付耀方
程涛
XIAO Jin-li;FU Yao-fang;CHENG Tao(School of Navigation,Wuhan University of Technology,Wuhan 430063,China;Hubei Key Laboratory of Inland Shipping Technology,Wuhan 430063,China)
出处
《大连海事大学学报》
CAS
CSCD
北大核心
2018年第4期8-14,共7页
Journal of Dalian Maritime University
基金
湖北省自然科学基金面上项目(2015CFB282)
关键词
单锚泊船舶
走锚监测
AIS信息
改进DBSCAN算法
single anchored ship
anchor monitoring
AIS information
improved density based spatial clustering of applications with noise (DBSCAN) method