ROME LABORATORY'S DRAFT FY98 SBIR TOPICS PART 7 OF 10. ROME LABORATORY'S DRAFT FY 98 SBIR TOPICS. ROME LABORATORY IS PLEASED TO MAKE AVAILABLE THE FOLLOWING DRAFT SMALL BUSINESS INNOVATIVE RESEARCH (SBIR) PROGRAM TOPICS. THESE TOPICS ARE NOT APPROVED AND ALL MAY NOT APPEAR IN THE FINAL SOLICITATION: SBIR TOPIC #AF98-128. TECHNICAL POINT OF CONTACT: Joseph P. Cavano, RL/C3CB (315) 330-4033. TITLE: Mobile, Adaptive Knowledge Base Decision Agents. CATEGORY: Research and Development. DOD CRITICAL TECHNOLOGY AREA: B08. SERVICE CRITICAL TECHNOLOGY AREA: AF1. OBJECTIVE: To develop collaborative software agents that can play an active role in decision-making while solving the challenges of a global, net-based C4I information environment. DESCRIPTION: Autonomous software components that are active, persistent and can reason and communicate while navigating heterogeneous computing environments are typically considered intelligent agents.' Most agents perform a specific task on behalf of a user but usually they do not interact with others in a collaborative paradigm. They also tend to concentrate on the first phase of decision-making (i.e., searching the environment for information). This topic seeks to build upon agent technology developed by DARPA and Rome Lab and their corresponding access to existing C4I knowledge bases and expand it into the second and third phases of decision-making inventing, developing and analyzing possible courses of action and then selecting, recommending and communicating the best course of action. Agents need to be accessible from anywhere on the globe and be able to communicate key findings and results to appropriate entities, whether human or other machine agents. Agents may also be lazy in that they do not have to perform all the work themselves; instead, they can take advantage of other agents, information servers or problem solvers. Finally, agents must be adaptable and capable of serving a hierarchy of military users. PHASE I: Identify and justify a multi-agent technology approach to the decision-making process. The approach must be platform neutral, globally available, and build upon sound commercial technology whenever possible. Provide a report describing the concept and architecture in detail and develop an archetype knowledge base decision agent. PHASE II: Build an initial system by constructing several such agents and evaluate their effectiveness in solving specific C4I problems. COMMERCIAL POTENTIAL: Phase III will develop a full-scale system that will commercialize the results of this Phase I and Phase II. Agent technology can augment human decision-making and build upon Internet information. It will have enormous applicability to military and commercial domains in the information age, ranging from law enforcement, finance, education, news delivery and health. KEYWORDS: Adaptive Information, Knowledge Bases, Collaborative Decision Making, Intelligent Agent. SBIR TOPIC #AF98-129. TECHNICAL POINTS OF CONTACT: Robert M. Flo, RL/C3AB,(315) 330-2334; Peter A. Jedrysik, RL/C3AB (315) 330-2158 ;Richard C. Metzger, RL/C3CA (315) 330-7652. TITLE: Data Visualization for Collaborative Wargaming and Battlefield Management of C4I System. CATEGORY: Research and Development. DOD CRITICAL TECHNOLOGY AREA: B07. SERVICE CRITICAL TECHNOLOGY AREA: AF1. OBJECTIVE: Develop innovative technologies for improving current data visualization and manipulation capabilities for collaborative wargaming and battlefield management of C4I systems and large plans to enhance the overall decision-making process. DESCRIPTION: At the height of a military conflict, there is an abundance of automated information that must be managed efficiently to increase the pace of combat operations, improve the decision-making process, and synchronize various combat actions. The ultimate goal is: Give the battlefield commander, and his support staff, access to all information needed to win the war. And, more importantly, give it to him when he wants it, where he wants it, and how he wants it thus assisting the commander in more timely, informed, decisions on modifications tothe present plan. Advancements in these C4I technology areas must be made in order for the tri-services to interact together effectively and with accurate information. The challenge of this effort is the development of innovative techniques and approaches which enhance collaborative interaction, information understanding, and visualization of wargaming and battlefield management scenarios in virtual environments and in interactive high resolution, group viewing display environments. These approaches may be compatible within the evolving COTS environment found throughout both commercial and military systems. PHASE I should identify current limitations and proposed innovative techniques which offer significant improvements over current state of the art data visualization and manipulation technologies. PHASE II should accomplish a prototype development and/or demonstration which incorporates and demonstrates the proposed Phase I enhancements. PHASE III DUAL USE APPLICATIONS: Phase III will focus on commercialization of corresponding enhancements. KEYWORDS: Data Visualization, Data Manipulation, Collaborative, Interaction, Information Understanding, Virtual Environments, Planning, Air Campaign. SBIR TOPIC #AF98-130. TECHNICAL POINT OF CONTACT: Dr. Raymond A. Liuzzi, RL/C3CA (315) 330-3528. TITLE: Dynamic Data Mining. CATEGORY: Research and Development. DOD CRITICAL TECHNOLOGY AREA: B08. SERVICE CRITICAL TECHNOLOGY AREA: AF1. OBJECTIVE: Investigate and develop dynamic and adaptive data mining techniques for designing, developing, and accessing large-scale data/knowledge bases for intelligent information systems. The goal is to produce a dynamic intelligent information system architecture consisting of innovative memory mechanisms, high performance intelligent agent supported architecture innovations and layers of very large data/knowledge bases capable of coordinating, cooperating and negotiating to provide just-in-time information and services. DESCRIPTION: Investigate high performance computational mechanisms to enhance data mining in massive high performance information data/knowledge bases to support joint efforts of DARPA and Rome Laboratory in technology innovation for the area of intelligent information systems. The growing diversity of different types of data is generating a problem because of the massive size of modern data/knowledge bases. Increased use of video, fax, graphics, images, voice, and textual data make these data types readily available, in different forms, to users. Advanced computational models need to address processing of data at very high speeds (petaops). Adaptive memory techniques in conjunction with advanced data structures could provide innovative ways to both access and store various forms of data/knowledge. Intelligent ways to coordinate various forms of raw data, including restructuring, to discover information, and using new computational paradigms available in emerging high performance computing technology need to be investigated. Optical computing innovations could provide breakthroughs in the area of special purpose architecture enhancements. Graphical tools and machine learning techniques for information discovery require both adaptable and scaleable innovations. Mobile computing designs offer potential for incorporating large information sources within local domains. Research innovations in these areas supporting dynamic data mining technology will help provide ways data should be dynamically structured and stored for efficient retrieval as well as provide adaptable transformation techniques to structure knowledge which can be managed more efficiently so that information can be automatically filtered, manipulated and summarized. Mechanisms to be investigated include (1) intelligent information rich hyperprogram web agents, (2) advanced adaptable memory design/configurations, (3) electro/optical special purpose architecture enhancements, (4) mobile hand-held computational mechanisms for seamless access, and (3) evolvable data/knowledge base configurations for scalable information aggregation/processing. Technical challenges include unique use of adaptive architectures, dynamic databases, and information integration. PHASE I will investigate development of techniques for designing, developing and integrating large-scale active information systems using massive multi-source data rich repositories. PHASE II will demonstrate a dynamic adaptable data/knowledge base configuration for very large knowledge bases in appropriate scalable information processing domains/platforms. PHASE III DUAL USE APPLICATIONS: Phase III will test dynamic information mining tools for rapid knowledge base access and commercialize results of Phase I and II. Rapid accessibility to integrated systems and information increases choices for consumers in both civilian and defense applications. This technology could have a major impact on applications that require integrated decision making and timely and accurate information such as planning/scheduling systems, autonomous vehicles, aircraft operation, hospital life support systems, decision support systems and personal military command and control. KEYWORDS: Intelligent Systems, Software, Adaptive Computers, Knowledge Base, Dynamic Data Base. Margot Ashcroft, SBIR Program Manager, RL/XPD, 315-330-1793, Joetta A. Bernhard, Contracting Officer, RL/PKPX, 315-330-2308.

Loren Data Corp. http://www.ld.com (SYN# 0605 19970701\SP-0009.MSC)