Scribd
Upload
279 views1 page

Beyond The Kalman Filter: Particle Filters For Tracking Applications

The document discusses particle filters as an alternative to the Kalman filter for target tracking applications. It reviews the development of particle filters and their advantages over Kalman filters for nonlinear or non-Gaussian problems. Several target tracking applications are described where particle filters may provide more accurate and reliable estimates than conventional Kalman based methods.

Uploaded by

arab
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
279 views1 page

Beyond The Kalman Filter: Particle Filters For Tracking Applications

The document discusses particle filters as an alternative to the Kalman filter for target tracking applications. It reviews the development of particle filters and their advantages over Kalman filters for nonlinear or non-Gaussian problems. Several target tracking applications are described where particle filters may provide more accurate and reliable estimates than conventional Kalman based methods.

Uploaded by

arab
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 1

Beyond the Kalman Filter: Particle Filters for Tracking

Applications1
Neil Gordon
Intelligence, Surveillance and Reconnaissance Division
Defence Science and Technology Organisation
PO Box 1500, Edinburgh SA 5111, Australia
[email protected]
Abstract - Target tracking is an important element of surveillance, guidance or obstacle avoidance, whose role is to determine the
number, position and movement of targets. The fundamental building block of a tracking system is a filter for recursive state
estimation. The Kalman filter has been flogged to death as the work-horse of tracking systems since its formulation in the 60's. In this
talk we look beyond the Kalman filter at sequential Monte Carlo methods, collectively referred to as particle filters. Particle filters
have become a popular method for stochastic dynamic estimation problems. This popularity can be explained by a wave of optimism
among practitioners that traditionally difficult nonlinear/non-Gaussian dynamic estimation problems can now be solved accurately
and reliably using this methodology. The computational cost of particle filters have often been considered their main disadvantage, but
with ever faster computers and more efficient particle filter algorithms, this argument is becoming less relevant. The talk is organized
in two parts. First we review the historical development and current status of particle filtering and its relevance to target tracking. We
then consider in detail several tracking applications where conventional (Kalman based) methods appear inappropriate (unreliable or
inaccurate) and where we instead need the potential benefits of particle filters.
1 The

paper was written together with David Salmond, QinetiQ, UK.

You might also like