Title :
Tracking maneuvering targets with an interacting multiple model filter containing exponentially-correlated acceleration models
Author :
Blair, W.D. ; Watson, G.A. ; Rice, T.R.
Author_Institution :
Weapons Control Div., Naval Surface Warfare Center, Dahlgren, VA, USA
Abstract :
An interacting multiple model (IMM) filter with constant velocity and exponentially-correlated acceleration (ECA) models is proposed for tracking maneuvering targets. An IMM filter is implemented separately in each Cartesian coordinate to allow for different model probabilities in each coordinate. For each coordinate, one of two ECA models is designed to track an increasing acceleration while the other ECA model is designed to track a decreasing acceleration. The IMM filter is designed so that the ECA model with increasing acceleration provides transition from the constant velocity model to the ECA model with decreasing acceleration. The ECA model with decreasing acceleration provides transition to the constant velocity model. Simulated tracking results are given for a constant speed target performing a `S´ trajectory maneuver
Keywords :
filtering and prediction theory; pattern recognition; picture processing; Cartesian coordinate; exponentially-correlated acceleration models; interacting multiple model filter; maneuvering targets; target tracking; Acceleration; Degradation; Differential equations; Filters; Integrated circuit modeling; Integrated circuit noise; Noise measurement; Target tracking; Time measurement; Trajectory;
Conference_Titel :
System Theory, 1991. Proceedings., Twenty-Third Southeastern Symposium on
Conference_Location :
Columbia, SC
Print_ISBN :
0-8186-2190-7
DOI :
10.1109/SSST.1991.138553