Title :
On Kalman, H∞ and H2 target tracking: probability of target escape
Author :
Rawicz, P.L. ; Kalata, P.R. ; Murphy, K.M. ; Chmielewski, T.A.
Author_Institution :
Drexel Univ., Philadelphia, PA, USA
Abstract :
This paper considers the Kalman, H∞ and H2 continuous time, steady state, target tracking problem. The target escape criteria is examined, in particular the probability that the position estimate will sufficiently degrade causing the target to be outside the radar beam width. Analytical expressions for the probability of target escape are developed not only to compare the performance of the various filters, but also as a design tool to meet tracking specifications. It is shown that the Kalman and H2 trackers have a lower probability of escape than the H∞ tracker when the target manoeuvrability is Gaussian and within the tracker design setting. Conversely, the H∞ process generally has a lower probability of escape than the Kalman and H2 trackers when the target manoeuvrability is a pulse acceleration
Keywords :
Kalman filters; filtering theory; probability; radar tracking; target tracking; tracking filters; H∞ filters; H2 filters; Kalman filters; probability; radar; target manoeuvrability; target tracking; tracking index; Equations; Filtering; Hydrogen; Kalman filters; Position measurement; Radar measurements; Radar tracking; State estimation; Steady-state; Target tracking;
Conference_Titel :
American Control Conference, 2000. Proceedings of the 2000
Conference_Location :
Chicago, IL
Print_ISBN :
0-7803-5519-9
DOI :
10.1109/ACC.2000.877044