|
8:
An Introduction to the Kalman Filter
Sunday, Tutorial, 10 am - noon
Room 404
The
40-year-old Kalman filter and related optimal estimators
continue to appear in a wide variety of computer graphics
applications, such as simulating musical instruments in
virtual reality, head tracking and motion capture, extracting
lip motion from video sequences of speakers, and fitting
spline surfaces over collections of points. The Kalman
filter is an optimal estimator for a large class of problems
and a very effective and useful estimator for an even
larger class. In its most basic form it is also relatively
simple to use and understand. This tutorial presents an
intuitive approach that enables developers to approach
the extensive literature with confidence.
Prerequisites
Basic linear algebra. A basic mathematical background
sufficient to understand explanations that involve introductory
statistics and random signals.
Topics
An intuitive explanation of the filter. The origins and
formulations of the filter equations. Practical use of
the filter. Approaches for non-linear systems. A brief
introduction of advanced topics such as sensor, information,
and data fusion; sensitivity and stability concerns; system
identification (tuning); multi-modal (multiple-model)
approaches; and optimal smoothing.
Organizers/Lecturers
Gregory Welch
Gary Bishop
University of North Carolina at Chapel
Hill |
|
