{PLEN/TTFMOSTvbABS+het 170120}


TARGET TRACKING AND DATA FUSION:

How to Get the Most Out of Your Sensors (and make a living out of this)

Yaakov Bar-Shalom, Distinguished IEEE AESS Lecturer
Recipient of the 2008 IEEE Dennis Picard Medal for Radar Technologies
University of Connecticut
Storrs, CT 06269-4157, USA
E-mail: ybs@engr.uconn.edu


This talk discusses the issues related to information extraction and data fusion from multiple sensors, in particular, from radars.
The goal of extracting the maximum possible amount of information from each sensor requires 
the use of appropriate sensor as well as  target models. In these models one has to quantify the corresponding uncertainties.
The issues related to data association and multiple target behavior models are discussed together with some practical algorithms a
nd their implementations for Low Observable targets, with an example of early acquisition of a VLO TBM.
The fusion of the information from various sources has to account for their uncertainties
as well as the interrelationship -- crosscorrelations -- between the uncertainties across sources.
The "Track-to-Track Fusion" and "Centralized Fusion" configurations are discussed.

The conventional fusion techiques have been developed for the case where the state vectors used at different sensors are the same.
However, the situation of different state vectors used at different - heterogeneous - sensors requires a special treatment.
While the centralized fusion is the best for linear systems with for homogeneous sensors, the recently developed 
fusion of state estimates from heterogeneous sensors shows that decentralized fusion can be superior to the centralized one.
This is illustrated on an example of fusion from a radar and and an ESM (or EO) sensor for a maneuvering target, 
which requires a nonlinear IMM estimator.


REFERENCES

Y. Bar-Shalom, P. K. Willett and X. Tian, "Tracking and Data
Fusion: A Handbook of Algorithms", YBS Publishing, 2011 (available from amazon.com).
T. Yuan, Y. Bar-Shalom and X. Tian, ``Heterogeneous Track-to-Track Fusion'', 
J. of Advances in Information Fusion}, 6(2):131--149, Dec. 2011.