An Adaptive Data Fusion and Analysis System

Award Information
Agency:
Department of Defense
Amount:
$99,906.00
Program:
SBIR
Contract:
N00421-04-P-0573
Solitcitation Year:
2004
Solicitation Number:
2004.1
Branch:
Navy
Award Year:
2004
Phase:
Phase I
Agency Tracking Number:
N041-016-0946
Solicitation Topic Code:
N04-016
Small Business Information
AMERICAN GNC CORP.
888 Easy Street, Simi Valley, CA, 93065
Hubzone Owned:
N
Woman Owned:
N
Socially and Economically Disadvantaged:
Y
Duns:
611466855
Principal Investigator
 Tasso Politopoulos
 Chief Scientist
 (805) 582-0582
 tpolito@americangnc.com
Business Contact
 Sandee Baker
Title: sbaker@americangnc.com
Phone: (805) 582-0582
Email: sbaker@americangnc.com
Research Institution
N/A
Abstract
This project develops an adaptive data fusion and analysis system to intelligently process, analyze, combine, and abstract useful knowledge from heterogeneous test data sources aboard aircraft/ship by integrating evidence theory, fuzzy logic, plausible reasoning, and dynamic probabilistic networks. Evidence theory serves as a rule induction and knowledge extraction process. A fuzzy clustering procedure extracts information from the original data sources. A plausible reasoning module permits the synthesis of knowledge acquired from different sources and performs causal inference processes to enhance decision aid processes. Dynamic probabilistic networks serve to maintain and update the belief state over time, providing update information for optimal decision making. The proposed intelligent information fusion and analysis system not only takes advantage of innovative individual methodologies but also, more significantly, emphasizes the synergism among all subsystems to assure the overall performance of the information fusion and analysis system. The end product is a commercial software package, An Adaptive Data Fusion and Analysis System, which will dramatically increase the productivity of organizations involved in data management, data mining, data fusion and analysis, and structure and pattern extraction from huge databases of test aboard aircraft/ship. The effectiveness of the proposed techniques is demonstrated on a fully integrated prototype module at the end of Phase I.

* information listed above is at the time of submission.

Agency Micro-sites

US Flag An Official Website of the United States Government