The paper addresses how to efficiently exploit the knowledge-base (KB), e.g. environmental maps and characteristics of the targets, in order to gain improved performance in the tracking of multiple targets via measurements provided by a ship-borne radar operating in a littoral environment. In this scenario, the nonhomogeneity of the surveillance region makes the conventional tracking systems (not using the KB) very sensitive to false alarms and/or missed detections. It is demonstrated that an effective use of the KB can be exploited at various levels of the tracking algorithms so as to significantly reduce the number of false alarms, missed detections, and false tracks and improve true target track life. The KB is exploited at two different levels. First, some key parameters of the tracking system are made dependent upon the track location, e.g., sea, land, coast, meteo zones (i.e., zones affected by meteorological phenomena) etc. Second, modifications are introduced to cope with a priori identified regions nit hi high clutter density (e.g. littoral areas, roads, meteo zones etc.). To evaluate the behavior of the proposed knowledge-based tracking systems, extensive results are presented using both simulated and real radar data
Knowledge-based system for multi-target tracking in a littoral environment
IEEE Transactions on Aerospace and Electronic Systems ; 42 , 3 ; 1100-1119
2006-07-01
4023674 byte
Article (Journal)
Electronic Resource
English
Littoral Tracking using Particle Filter
British Library Conference Proceedings | 2002
|Littoral tracking using particle filter
IEEE | 2002
|Radar for the littoral environment
British Library Conference Proceedings | 1998
|Naval radar in a littoral environment
Tema Archive | 2004
|