Research On Fusion Method Of AIS And Radar Objects Location Data

In the foreseeable future, the Automatic Identification System(AIS) will be installed on all sea-going ships larger than 300g.r.t. These new devices broadcast data at regular and short intervals, with which the ship can be identified and tracked. Further data can be transmitted to provide additional information on the ship's motion, navigational state, intended voyage and cargo. The transmissions can be received by every interested party having an appropriate receiver both at sea and ashore. With these functions, AIS provides new, accurate and detailed information to maritime traffic, which is relevant not only to ship/ship but also to VTS/ship interaction. What's more, AIS can provide many functions that radar cannot provide. It makes up the shortages of radar in vessel collision avoidance. But, AIS can't replace radar in spite of the benefit it provides.AIS has its virtues as a means for surveillance and tracking of those ships that are equipped. Obviously, ships not equipped, or with a switched-off system, remain invisible, additionally, coastlines and fixed or floating navaids visible only if they be equipped with AIS. Because of their different characteristics, radar and AIS complement each other, so that the implementation of AIS provides considerable 'added value' to the existing state-of-the-art. Therefore, the united using of this two sensors will be a trend to assure navigation safety in the near future.The study in this field is just beginning both in the world and in our country. There are papers studied on the composition of AIS and ECDIS and the fusion method of object position information in radar and AIS etc. All of them are the beginning studies in the field, and it needs more devotion to boost the development of AIS.This paper studies at the data fusion of AIS and radar objects. AIS and radar objects' data fusion is composed of two important parts. The first is AIS and radar tracks correlation; the second is AIS and radar objects' location data synthesize. The former research on tracks correlation has been very mature. But there is few study on the AIS and radar tracks correlation. So this paper adopts some former methods of tracks correlation to resolve the problem of AIS and radar objects'tracks correlation. As to the AIS and radar objects' location data synthesize, the adopted method in previous studies of this field is usually simple Data Fusion method. They usually think little of the precision and practicability of the model. Considering of it, this paper puts forward a new method, which can realize both precision and practicability. Based on the data-fusion theory it united BP Neural Network algorithm and power-adding algorithm formed a data synthesize model.Simulation is giving in the end to test the result's precision and practicability of these two models. It proves that the two models are all practicable, robust and can get accurate data results, can be embedded to other bigger systems directly.