NASA SBIR 2003 Solicitation


PROPOSAL NUMBER:03-F2.02-9170 (For NASA Use Only - Chron: 033835)
SUBTOPIC TITLE:Multi-agent and Human-centric Systems Technologies
PROPOSAL TITLE:Agent-based Health Monitoring System

SMALL BUSINESS CONCERN (Firm Name, Mail Address, City/State/Zip, Phone)
Intelligent Automation, Inc.
7519 Standish Place Suite 200
Rockville ,MD 20855 - 2785
(301) 294 - 5200

PRINCIPAL INVESTIGATOR/PROJECT MANAGER (Name, E-mail, Mail Address, City/State/Zip, Phone)
Leonard S Haynes
7519 Standish Place Suite 200
Rockville ,MD  20855 -2785
(301) 294 - 5250
U.S. Citizen or Legal Resident: Yes

IAI has successfully used principal component analysis to detect even subtle latent faults in a subsystem. An advantage of the PCA for this application is that the system can be real-time, even when there are very large data sets to be analyzed. IAI has also successfully used software autonomous agents to provide robust, flexible, optimized control of large complex decentralized systems such as air traffic control, battle management, etc. The innovation of this proposal is the integration of these previously separate technologies to provide system health management and self-reliant systems. The system to which we will apply this innovation is a power grid where nodes can be power producers, power consumers, power transmission elements, and in some cases nodes can be either producers or consumers at different times. Our software agents interact with each other based on a contract net paradigm. IAI has been working on agent-based systems for a decade and has an extensive library of tools and techniques to design and implement agents interacting via a contract net. These tools and techniques will be used in the proposed work and will allow us to implement a simulation of a non-trivial system even during Phase 1.

The immediate application of this technology for NASA includes the ISS, the Space Shuttle, and the eventual replacement for the Space Shuttle. It can be applied directly to the power management systems therein, but it can also be more broadly applied to integrated health monitoring and control under fault conditions.

Our goal for the first commercial application of this technology will be use in the national power grid to avoid the type of blackout that darkened almost the entire northeastern part of the US. We believe that both the Government and power companies see a high priority in upgrading power management technology to avoid such massive failures, and we also believe that we have the most viable solution to that problem. Additionally the same agent framework can provide optimization of existing and future power generation, including micro-generation by businesses and even by individuals.