Title :
Vessel trajectory partitioning based on hierarchical fusion of position data
Author :
Xianbin Wu;Lin Wu;Yongjun Xu;Zhulin An;Boyu Diao
Author_Institution :
Institute of Computing Technology, Chinese Academy of Sciences, Beijing, China
fDate :
7/1/2015 12:00:00 AM
Abstract :
Maritime situational awareness now greatly benefits from AIS (Automatic Identification System). The most fundamental function a maritime situational awareness system should provide is vessel trajectory analysis. Previous work has shown the necessity of vessel trajectory partitioning before further analysis, but limited work existed for this problem. With billions of AIS messages received every month, computational complexity should be concerned as well as preciseness, and there is room for both of them to improve. This paper proposes a vessel trajectory partitioning method based on hierarchical fusion of position data reported by AIS, aiming at improving preciseness and speed. Our method consists of two levels: position fusion and sub-trajectory fusion. Experimental results based on real data demonstrate that our method splits routes and identifies stops precisely with the computational complexity of O(n).
Keywords :
"Trajectory","Aggregates","Marine vehicles","Partitioning algorithms","Ports (Computers)","Time complexity"
Conference_Titel :
Information Fusion (Fusion), 2015 18th International Conference on