Advanced Visualization and Support Environment (ADVISE)

An operator within the JSpOC faces a continuous and rapid flow of intelligence from disparate and distributed sensors, theatre participants, and other operators, all of which may or may not accurately reflect the current operational situation. Despite having technology that fuses these heterogeneous data sources, this abundance of information does not guarantee mission success or even improved SSA; rather it can overload JSpOC operators, resulting in confusion, poor analysis, and ineffective decision-making. To improve the accuracy, efficiency, and usefulness of these fusion algorithms employed within the JSpOC, we present our proposed effort to design and demonstrate a Framework Leveraging Operator Reasoning in Data Association and Filtering Algorithms (FLORIDA) to support improved SSA. First, we will perform a systematic evaluation of the JSpOC work domain to understand operator reasoning and decision-making requirements. Second, we will establish a meta-information ontology facilitating the augmentation of existing and future fusion algorithms with qualitative metrics. Third, we will select and augment several computation techniques currently employed within fusion algorithms. Fourth, we will design preliminary controls and visualizations that facilitate operator interaction with the FLORIDA-enhanced multi-source data fusion algorithms. Finally, we will develop and evaluate a demonstration prototype incorporating the above elements. BENEFIT: We see considerable promise in the commercial application of rapidly prototyped information and meta-information display and control techniques, specifically in the financial industry, where investment decision-making is fraught with uncertainty, and risk management is critical to investment strategy. We also see applications in commercial space asset management environments, where data quality and data recency are critical to asset protection. We also plan to transition the information and meta-information interface techniques developed with FLORIDA into the next version of our BNet® Bayesian belief network application.