Award Data

We propose a new automatic target recognition (ATR) system that operates on video data, exploiting the inherent redundancy and spatio-temporal coherence of motion imagery to create three-dimensional target signatures. Unlike previous approaches that useshape alone, our method will additionally model and predict object surface appearance properties, in the form of bi-directional reflectance distrib ...

Detecting and disrupting a terrorist network requires the ability to: link disparate (noisy, deceptive) data and recognize patterns in the underlying networks, estimate where further intelligence would be useful, and predict the network's response topossible attacks.We propose to combine state-of-the-art social network analysis (meta-networks) and statistical inferencing (probabilistic relational ...

Organic Air Vehicles have the potential to dramatically improve knowledge of the battlespace by providing persistent networked coverage which can dynamically adjust to changing conditions. These systems require novel algorithms which can address theproblems of positioning, sensor resource allocation, and tracking. Here we propose to define an architecture and create core algorithms which address ...

To predict likely terrorist actions and terrorist targets in the U.S. we need (1) models of how terrorist organizations evolve and learn (2) identification of cultural variables likely to shape the evolution of terrorist organizations within the U.S. and(3) methods for including these cultural context variables in organizational learning models. Our team includes three essential types of expertise ...

Bluefin Robotics proposes (in Phase I) to create a methodology for developing the Bluefin Agent Behavior Language (BABL) and associated tools. BABL will enable the rapid design and development of behaviors for autonomous vehicles. Key features of BABL arethat it:1) Makes a minimal set of assumptions about the capabilities and assumptions of the target UVs.2) Maintains generality without losing the ...

As the number of mobile source emissions in the United States increases, it is necessary to further reduce the emission of CO, NOx, and hydrocarbons from such sources to provide safe air quality, especially in urban environments. Significant reductions in pollution emissions can be achieved using advanced engine controls based on real-time measurements of CO, NOx, and hydrocarbon concentrations i ...

Currently available equipment for detecting submarines shows poor performance in littoral waters. We propose to deploy a population of inexpensive devices that will augment the capabilities of the currently available equipment to achieve the goal ofdetecting submarines in littoral waters. The U.S. military will achieve the capability of detecting submarines in littoral waters.

A number of very high-volume liquid chemicals form azeotropes with water and can be dehydrated to required purity levels only through the use of entrainers or drying agents. The handling and disposal of these additional chemicals present significant environmental risks. Recently, membrane pervaporation using both polymeric and inorganic membrane modules has been commercialized as a method to deh ...

Interpretation of video imagery is the quintessential goal of computer vision. The ability to group moving pixels into regions and then associate semantic labels with those regions has long been studied by the vision community. Only recently have thecomponent technologies matured sufficiently to make this goal attainable for well-defined scenarios. We propose a system for semantic interpretation o ...

Combined sewage outflows (CSOs) are a major contributor to water pollution in the United States. The impact of peak stormwater loads on sewage treatment plants leads to the discharge of significant amounts (1.26 billion gallons) of untreated sewage each year. The bacterial load represented by this discharge directly contributed to more than 2,200 beach closings in the year 2000, numerous closing ...