Research On Navigation Radar And AIS Information Fusion Method And Application

With the implement of maritime power strategy,the number of ships sailing in our sea is increasing greatly and Chinese maritime traffic conditions has become more complicated.Therefore,obtaining higher quality information for the coast guard law enforcement vessel during patrols is becoming more and more demanding.The coast guard law enforcement ship obtains information of surrounding ships through navigation radar and AIS(Automatic Identification System)equipment.It’s difficult for a single sensor to react towards the changing situation.Integrating the information acquired by each sensor brings about great significance of improving the accuracy,timeliness and comprehensiveness as well as the regulation of maritime traffic and the protection of maritime rights and interests.In order to provide a reliable maritime intelligence support for a coast guard law enforcement ship,this dissertation focuses on information fusion of navigation radar and AIS,follows the fusion model of distributed multi-sensor information,studies the track correlation and track fusion algorithm,and completes the design and implementation of fusion system.The main research works are as follows:1.Constructing an overall guidance framework for navigation radar and AIS information fusion,analyzing the fusion of navigation radar and AIS information,and establishing an information fusion model.2.Based on the research of decision tree classification algorithm,a track correlation model based on decision tree has been constructed.The decision tree is used to judge whether the navigation radar target is associated with the AIS target from the same time,and the historical relevance is updated.Then,the most likely relevant target based on the historical relevance will be verified and later can make the final correlation decision.3.Aiming at the problem of navigation radar and AIS track fusion,a track fusion model combining statistical weighting and extended Kalman filtering is constructed.The optimal weight is calculated based on the minimum mean square error criterion,and the track is integrated according to the statistical weighting method.Then,the extended Kalman filter is performed on the weighted fusion track to obtain the fused system track.4.Carrying out the simulation experiment of the navigation radar and AIS information fusion method.Designing training samples to generate the decision tree model;Designing testing samples to verify the correctness of the track correlation.Then,based on results of track correlation,the tracks of the relevant targets are fused.5.Combining with the actual requirements of navigation radar and AIS information fusion,design the basic fusion software function module of application system for test.The typical sea multi-objective application scenarios of information fusion are used to illustrate the feasibility of the functional module.The simulation case shows that the information fusion method studied in this dissertation has certain referential significance for improving multi-objective information fusion effect at sea and reducing the estimation error of the target state.