|PROPOSAL NUMBER:||04 T1.01-9963|
|RESEARCH SUBTOPIC TITLE:||Information Technologies for System Health Management, Autonomy and Scientific Exploration|
|PROPOSAL TITLE:||Hierarchical Intelligent Data Fusion Architecture for System Health Management|
|SMALL BUSINESS CONCERN (SBC)||RESEARCH INSTITUTION (RI)|
|NAME:||Global Technology Connection Inc||NAME:||Georgia Tech|
|ADDRESS:||2839 Paces Ferry Road, Suite 1160||ADDRESS:||505 Tenth Street NW|
PRINCIPAL INVESTIGATOR/PROJECT MANAGER
(Name, E-mail, Mail Address, City/State/Zip, Phone)
Dr. Freeman Rufus
TECHNICAL ABSTRACT (LIMIT 200 WORDS)
The complexity of modern systems and the stringent performance requirements for operation and uptime suggest that optimum and robust means must be deployed to make effective use of multiple sensor suites for assessing risk, identifying system degradation, understanding how system degradation progresses to failure, etc. Global Technology Connection and Georgia Tech proposes the development of data fusion architecture based on a hybrid analytical / intelligent methodology that exploits the concept of "focus of attention" via active perception in order to optimize degradation/fault classification accuracy while reducing substantially the computational burden. The fusion scheme incorporates several levels of abstraction: fusion at the data level, the feature level and the sensor level. The overall architecture employs technologies from soft computing, Dempster-Shafer theory and game theory to provide a robust and reliable platform for critical aerospace systems. Phase I effort will develop a data fusion algorithms for system degradation/fault identification.
Phase II will address design and construction of prototype field hardware for implementing the data fusion concept for components.
Several aerospace end users like Lockheed Martin and Boeing have already expressed interest in the commercial applications (Phase III) of this approach for health monitoring and life determination of Aerospace vehicles/systems.
POTENTIAL NASA COMMERCIAL APPLICATIONS (LIMIT 100 WORDS)
The results of this project will assist NASA initially and other government or industry sectors eventually to fully deploy robust and reliable identification of system degradation/fault from multiple sources of raw data. When integrated into an intelligent engine environment, such practices will reduce substantially the risk of false ID of faults while achieving fault detection in the shortest possible time, increase the availability of aircraft systems and improve their reliability in the execution of critical missions. The aerospace, air traffic control, transportation, biomedical and other industries will also benefit from the introduction of these "smart" technologies into their operations.
POTENTIAL NON-NASA COMMERCIAL APPLICATIONS (LIMIT 100 WORDS)
Detecting and identifying fault and damage in aerospace systems for DoD and other government / private industries: air traffic control, pump manufacturers, HVACs, transportation, biomedical, etc.