NASA SBIR 2004 Solicitation


SUBTOPIC TITLE:On-Board Science for Decisions and Actions
PROPOSAL TITLE:Autonomous Collaborative Agents for Onboard Multi-Sensor Re-Targeting

SMALL BUSINESS CONCERN (Firm Name, Mail Address, City/State/Zip, Phone)
Intelligent Automation, Inc.
15400 Calhoun Drive, Suite 400
Rockville ,MD 20855 - 2785
(301) 294 - 5221

PRINCIPAL INVESTIGATOR/PROJECT MANAGER (Name, E-mail, Mail Address, City/State/Zip, Phone)
Goutam   Satapathy
15400 Calhoun Drive, Suite 400
Rockville, MD  20855 -2785
(301) 294 - 5249

TECHNICAL ABSTRACT (Limit 2000 characters, approximately 200 words)
In our Phase I effort we developed a prototype software-agent based framework to provide for autonomous re-targeting of sensors hosted on satellites in polar orbits, subject to standing requests and dynamic user requests. Our approach is a distributed and de-centralized, with each satellite maintaining knowledge of its sensor complement and physical capabilities as well as its schedule. Each satellite is capable of entering into negotiations for a viewing opportunity via a Contract Net type protocol; and is capable of optimizing its bid. As sensor deployment increases and autonomous vehicles become more robust, the need for a framework to support multi-sensor re-targeting in a dynamic environment involving multiple organizations becomes more acute. In our Phase 2 effort, we will extend the framework to include support for multiple types of sensor platforms as well as more sophisticated variants of the Contract Net Protocol. We will define various agent 'templates' specialized to the needs of the multi-sensor re-targeting environment. In Phase 2, the improved framework, AAMSRT, will support policy utilization to govern the sensor re-tasking. As part of the Phase 2 effort, we will engage the Open Geospatial Consortium for applicable standards, including use of SensorML, services, and OGC initiatives regarding sensor planning.

POTENTIAL NASA COMMERCIAL APPLICATIONS (Limit 1500 characters, approximately 150 words)
Commercialization potential of our proposed de-centralized and distributed AAMSRT framework extends to those cases involving a dynamic situation in which numerous sensor resources hosted on different types of platforms with different capabilities must be leveraged. This situation occurs in a number of application areas; hence, these areas hold the largest commercialization potential. The overarching application area area for NASA is that of 'Sensor Web', which involves specific applications such as earth observing and remote sensing projects for ecology applications and basic science.

POTENTIAL NON-NASA COMMERCIAL APPLICATIONS (Limit 1500 characters, approximately 150 words)
Other application areas for our proposed de-centralized and distributed AAMSRT framework involve homeland security, commercial industry security, and DoD applications. One homeland security need occurs at land borders, involving fixed sensor assets, mobile robots, and UAVs. Commercial industries also have the need for multi-sensor retargeting further with regard to sprawling critical infrastructure areas, such as in rail yards; this involves fixed sensor nets and sensors located on mobile robots. The US Coast Guard has a similar need for AAMSRT; this involves sensors that oversee harbor areas, including sensors that are hosted on surface vessels and sensors that are located in buoys and on wharfs. DoD military applications represents yet another area for AAMSRT framework usage.

Form Printed on 08-01-05 13:52