基于时空约束和三角形迭代划分的渔船AIS与ARPA轨迹匹配

AIS trajectory and ARPA trajectory matching of fishing vessels based on temporal-spatial constraints and triangle iterative partitioning

针对现有的轨迹相似度匹配算法用于渔船AIS(Automatic Identification System)与ARPA(Automatic Radar Plotting Aid)轨迹匹配时存在复杂度高、效率低等问题,本文提出基于时空约束和三角形迭代划分的渔船AIS与ARPA轨迹匹配算法TSC-TIP(Temporal and Spatial Constraint-Triangle Iterative Partitioning)。首先采用时空约束法筛选出ARPA目标时空约束范围内的AIS数据;其次采用三角形相似算法选择与AIS数据具有相似特征点的APRA轨迹数据;最后设计了子轨迹迭代划分法将每条轨迹划分为两条子轨迹并采用三角形相似法对子轨迹进行迭代筛选。为验证算法的性能,用渔船的真实AIS轨迹数据和ARPA轨迹数据进行了试验,结果表明:与基于经典距离的相似性度量方法相比,提出的TSC-TIP算法在不影响匹配准确率的前提下,匹配时间减少了95%。研究表明:TSC-TIP算法能有效匹配渔船AIS与ARPA轨迹数据,为面向AIS与ARPA的渔船轨迹数据融合研究提供了新思路。

The existing trajectory matching algorithms are high complexity and low efficiency for AIS(Automatic Identification System)trajectory and ARPA(Automatic Radar Plotting Aid)trajectory matching of fishing vessels.In order to solve this problem,a trajectory matching algorithm TS-TIP(Temporal and Spatial Constraint-Triangle Iterative Partitioning)based on temporal spatial constraint and triangle iterative partitioning is proposed.Firstly,the temporal and spatial constraintalgorithm is applied to filter the AIS data within the same spatio-temporal area with the ARPA data.Secondly,the triangle similarity algorithm is used to select the APRA trajectories including the same feature points with the AIS data.Finally,sub-trajectories iterative partitioningalgorithm is designed to determine the matched trajectory in which each trajectory is divided into two sub-trajectories and the sub-trajectories are selected iteratively with the triangle similarity algorithm.The experiments on real AIS trajectory data and ARPA trajectory data are designed to test the performance of the proposed algorithm.The results show that the matching time of the TSC-TIP algorithm reduces 95%compared with that of the classical distance-based similarity measurement method without affecting accuracy.The research indicates that the TSC-TIP algorithm can effectively match AIS trajectory and ARPA trajectory,and also provide new ideas for the data fusion on AIS and ARPA of fishing vessels.