Target Tracking in its simplest possible description can be defined as:

"Target Tracking is the prediction of the future location of a dynamical system based on its  estimates and measurements."

Analyzing this statement, one shall critically raise questions like: "What other possible information could be obtained by the target tracking?", "How do we model the target dynamics and how is it updated with measurements?", "What kind of measurements do we receive?", "Do we have other useful information available?", etc., etc. The list of questions could be continued depending on the specific environment for which the tracking algorithm is designed. A very important question is the number of targets being tracked since various issues arise for multiple target tracking (MTT). The single target tracking (STT) system is widely used to redirect the sensor pointing angle to follow the target, whereas the MTT covers a certain area commonly called the field of view (FOV). Within the FOV several sensor can receive measurements at irregular intervals, which are required to be assigned to the possible multiple targets by a complex date association logic. Initially developed track-while-scan (TWS) systems for STT have been extended for the MTT application as described by Hovanessian. Modern MTT techniques combine data association and Kalman filtering as developed by Bar-Shalom and Singer.

The topic of target tracking involves a broad area of research interests, such as:

Preceding the above, a precise definition of target tracking found in "Design and Analysis of Modern Tracking Systems" by Samuel Blackman and Robert Popoli, 1999 states as:

    "The target tracking objective is to collect sensor data from a field of view containing one or more potential targets of interest and to then partition the sensor data into sets of observations, or tracks , that are produced by the same sources. Once tracks are formed and confirmed (so that background and other false targets are reduced), the number of targets can be estimated and quantities, such as target velocity, future predicted position, and target classification characteristics, can be computed for each track." (qtd. in Blackman, Popoli 1999)

These web pages are intended to contribute to the tracking community with the development of novel methods in target tracking and as a source of applications and references.

Content last reviewed: 28 Jul 2003 11:10:33 -0400 December 06., 2002