AVID: Agent for Visualization and Intelligent Decision-Aiding

We propose to develop a robust, adaptive Agent for visualization and Intelligent Decision-aiding (AVID). The AVID architecture consists off four modules, one each for: decision-aiding, visualization, profile adjustments, and communication and coordination. the structure of the key decision-aiding and visualization modules closely matches a unified cognitive theory, so as to provide assistance at multiple levels of information processing: skill-based, rule-based, and knowledge-based The decision-aiding module is the core component of AVID and provides three key decision-support functionalities: 1) fusion of data into task-relevant information, supporting skill-based processing; 2) event detection and situation assessment providing high-level aggregates of task-relevant knowledge, supporting rule-based processing; and B) response recommendations based on current context and prior knowledge, supporting knowledge-based processing. Robust performance is assured through the use of complementary AI techniques (e.g., fuzzy logic, belief nets, probabilistic rule-bases), and by providing aiding at the highest level allowed by available data and knowledge. Adaptive assistance is assured by allowing the user to customize an information and display profile to best fit the current situation. We propose to develop AVID using COTS software, and demonstrate it in the domain of battlefield monitoring. BENEFITS: Commercial applications of the generic AVID architecture exist in a variety of contexts characterized by high-information flow rates, decision-making under uncertainty, and distributed players, cial services, rail and air traffic control. Vertical applications for these markets would significantly improve effectiveness and safety in a number of life-critical applications.