Intelligent Information-Gathering Agents
Agency / Branch:
DOD / NAVY
Toyon Research Corporation proposes to develop a novel algorithm for dynamically prioritizing information available to members of a distributed intelligence architecture. This algorithm will allow a network agent to determine what information is of most value to client agents onboard aircraft based on tactical objectives and the current battlefield situation. Our approach is to use Dynamic Bayesian Networks (DBNs) - a leading technique in the field of artificial intelligence - as a data structure to model the current information content of each agent in the network and determine the appropriate level of detail to transmit. As new data is received by an agent in the network, the resulting improvement in mission-specific knowledge is quantified through measures from information theory. If the change is significant, the network agent transmits the information to appropriate agents in the network. In Phase I, we will develop appropriate DBNs for representative naval missions, develop a software prototype, and demonstrate the feasibility of the approach and algorithms. In Phase II, we will refine the scheme and develop a more sophisticated prototype that would allow us to demonstrate and evaluate the scheme.
Small Business Information at Submission:
TOYON RESEARCH CORP.
6800 Cortona Drive Goleta, CA 93117
Number of Employees: