A concept of decentralized fusion of maritime radar targets with multisensor Kalman filter

The paper presents a concept and an algorithm of multisensor decentralized data fusion for radar tracking of maritime targets. The fusion is performed in the space of Kalman Filter and is done by finding weighted average of single state estimates provided be each of the sensors. The article presents both algorithms - Kalman Filter for tracking objects in single sensor and combining them together to find one fused state vector. Another approach for target tracking, namely neural target tracking is also recalled in the aspect of fusion. Two approaches for data fusion - centralized and decentralized - are stated and the latter is thoroughly examined. The discussion on main problems involved in fusing process in complex radar systems is then presented. This includes coordinates transformation, track association and measurements synchronization. Future plans of including neural tracking in data fusion are presented. The article is ended with summary of the issues pointed out in it.