Advanced Multisensor Fused Track and Discrimination Architecture

Award Information
Agency:
Department of Defense
Branch:
Missile Defense Agency
Amount:
$749,920.00
Award Year:
2007
Program:
STTR
Phase:
Phase II
Contract:
HQ0006-07-C-7651
Agency Tracking Number:
B064-003-0054
Solicitation Year:
2006
Solicitation Topic Code:
MDA06-T003
Solicitation Number:
N/A
Small Business Information
DECIBEL RESEARCH, INC.
PO Box 5368, Huntsville, AL, 35814
Hubzone Owned:
N
Socially and Economically Disadvantaged:
N
Woman Owned:
N
Duns:
121016096
Principal Investigator
 Enrico Poggio
 Chief Technical Officer
 (256) 489-6124
 ecp@dbresearch.net
Business Contact
 Eric Cochran
Title: Chief Financial Officer
Phone: (256) 489-6125
Email: eric@dbresearch.net
Research Institution
 ALABAMA A&M UNIV. RESEARCH INS
 Kevah Heidary
 4900 Meridian Street
PO Box 313
Normal, AL, 35762
 (256) 372-5587
 Domestic nonprofit research organization
Abstract
This effort will extend the Phase I design, development, implementation, testing and performance evaluation of an automated prototype Multisensor Attributes and Contextual Information-Aided Track Correlation Algorithm. The proposed algorithm integrates the following modules based modified/enhanced existing stand-alone algorithms: a) A basic Multisensor Data Collection and Feature Extraction Module, b) A Data and Feature Fusion Engine that combines data and features from multiple sensors and automatically provides object dynamic motion histories, c) a Features-to-Attribute conversion module that automatically takes the data from the Data and Feature Fusion Engine and converts Features into Attributes, d) a Contextual Information Capture Algorithm based on, and e) an Enhanced Object Hypothesis Testing Module based on a new Cost Function Criteria for a joint Metric-Attributes-Contextual Information Track Correlator. The addition of contextual information is expected to enhance the algorithm performance in cases when: 1) the Features-to-Attributes process may be degraded due to insufficient data collection and/or poor feature extraction results, 2) two or more hand-over sensors have different numbers of objects in their field of view. The automated prototype algorithm will undergo Design-to-Capability testing at the MDA/DVH AMP facility and, as an option, it would be tested in one of the BMDS Fusion Testbed "Sidecar" processors.

* information listed above is at the time of submission.

Agency Micro-sites

US Flag An Official Website of the United States Government