Title :
Extending the IMM filter to Heterogeneous-Order state space models
Author :
Lopez, Rémy ; Danés, Patrick ; Royer, François
Author_Institution :
LAAS, CNRS, Toulouse, France
Abstract :
An extension of the celebrated Interacting Multiple Model (IMM) filter is proposed so as to handle a bank of models with state vectors of heterogeneous size and meaning. The derivation of the IMMHO (IMM Heterogeneous-Order) filter is exposed. It is argued that a component-wise cooperation strategy leads to a sound and versatile solution. Simulation results in the context of satellite telemetry, for which the prior dynamics of the moving target is poorly known, show that the filter is well-behaved, with adequate operation across the various modes. All the theoretical background is provided so that the paper is self-contained.
Keywords :
Kalman filters; satellite telemetry; target tracking; IMM Kalman filter; IMM heterogeneous-order filter; component-wise cooperation; heterogeneous-order state space model; interacting multiple model; moving target tracking; satellite telemetry; state vector; Approximation algorithms; Approximation methods; Equations; Kalman filters; Markov processes; Mathematical model; Satellites; IMM Kalman filters; IMMHO filter; heterogeneous-order switching models; target tracking;
Conference_Titel :
Decision and Control (CDC), 2010 49th IEEE Conference on
Conference_Location :
Atlanta, GA
Print_ISBN :
978-1-4244-7745-6
DOI :
10.1109/CDC.2010.5717415