ENHANCED VARIABLE DIMENSION FILTER FOR MANEUVERING TARGET TRACKING

An enhancement of the variable dimension (VD) filter [1] for maneuveri ng target tracking is presented. The use of measurement concatenation, a procedure whereby fast sampled measurements are stacked while maint aining their proper relationships with the states, leads to significan t reduction in estimation error by low processing rate algorithms. The use of double decision logic (DDL) for the maneuver onset and ending detection as well as appropriate procedures for reinitialization of th e estimation filters result in improved maneuver detection and filter adaptation. Simulation results show the performance of the proposed en hanced variable dimension (EVD) filter.