基于样条卡尔曼算法的AIS数据修复  

Research of AIS Data Restoration Based on Spline Interpolation Kalman Algorithm

作　　者：苏俊杰 兰培真[1,2] SU Junjie;LAN Peizhen(Maritime Traffic Safety Institute,Jimei University,Xiamen 361021,China;National Engineering Laboratory for the Emergency Information Technology of Traffic Safety,Xiamen 361021,China)

机构地区：[1]集美大学海上交通安全研究所,福建厦门361021 [2]交通安全应急信息技术国家工程实验室,福建厦门361021

出　　处：《集美大学学报（自然科学版）》2022年第6期524-530,共7页Journal of Jimei University：Natural Science

摘　　要：针对船舶自动识别系统(automatic identification system,AIS)的异常数据修复问题,提出一种样条卡尔曼(spline Kalman,SK)算法。该算法根据船舶动力学原理,构建反映运动特征变化约束关系的系统状态转移模型,并以样条插值得到的AIS修复数据作为卡尔曼滤波器的观测数据,进而实现AIS数据的精确修复。采用厦门港及附近水域的历史AIS数据检验SK算法的有效性,检验结果表明:对于低缺失率的AIS数据集,SK算法的修复效果与样条插值算法相近,均优于KNN、RF和SVM算法,但随着AIS数据集缺失率的上升,只有SK算法具有较好的修复稳定性。该研究成果可以更加有效地修复AIS的异常数据,从而为海事大数据分析及相关应用提供良好的数据基础。In this paper,a spline Kalman(SK)algorithm is proposed for the repair of abnormal data of automatic identification system(AIS),which constructs a system state transfer model for reflecting the constraint relationship of the change of motion characteristics according to the ship dynamics,and uses the AIS restoration data obtained through spline interpolation as the observation data of the Kalman filter to realise the accurate restoration of AIS data.In this paper,the effectiveness of the SK algorithm is tested using the historical AIS data of xiamen port and nearby waters.The test results show that for the AIS data set with low missing rates,the restoration effect of SK algorithm is similar to that of spline interpolation algorithm,and all of them are better than KNN,RF and SVM algorithms,but as the missing rate of AIS data set increases,only SK algorithm has better restoration stability.It can be seen that the research results of this paper can repair abnormal data of AIS more effectively,thus providing a good data base for maritime big data analysis and related applications.

关 键 词：AIS 数据修复 三次样条插值 卡尔曼滤波 

分 类 号：U644.14[交通运输工程—船舶及航道工程]