Agents In Decision Aiding Systems (AIDA System)

Period of Performance: 01/30/2006 - 01/30/2008


Phase 2 SBIR

Recipient Firm

Toyon Research Corp.
6800 Cortona Drive Array
Goleta, CA 93117
Principal Investigator


Toyon proposes to further develop an intelligent agent-based system to provide expert assistance for the analyst in autonomously fusing, assessing, and collating relevant information. Agents will be expected to perform a variety of cooperative missions such as persistent intelligence, surveillance, and reconnaissance (ISR), target search, acquisition, and tracking, area denial, attack, route planning, and battle damage assessment under dynamically-varying bandwidth limitations, intermittent communications, and limited information. The system architecture provides a User-Interface Agent capable of communicating information to and from the User, including interfacing with existing combat applications. The system will be expanded to solve problems related to Homeland Security and distributed network centric warfare. The proposed architecture features centralized and decentralized solutions based on the information rate available to each agent within the cooperative network. The performance of the system will be guaranteed to degrade gracefully with the level of information available to each agent. The system comprises two data mining Agents, referred to as the Task Agent and the Information Agent. The Information Agent handles the details of communication with specific database formats that hold a history of sensor detections and track information. This data is then utilized by the Task Agent to fuse the data received from the Information Agent and create relevant situational awareness information that is then passed to the User-Interface Agent via a customizable Network Interface. Toyon will customize the software and hardware of the system to satisfy the needs of the Army and will use SLAMEM to generate real-world data sources and, in addition, will demonstrate the interfacing capability of the system by testing over a dynamic network environment suitable for a given scenario of interest.