Defense and Security
ARI works with key agencies throughout the Department of Defense and the Department of Homeland Security to build prototype solutions in a variety of key areas. These areas include advanced materials, algorithms for uncertainty quantification, track-stitching and machine learning techniques for autonomous systems, and sensor fusion.
ARI researchers are working to develop and process advanced materials to provide solutions to the Department of Defense in the areas of protective coatings, novel multifunctional nanocomposites, and field-deployable biosensors.
Spacio-temporal model of Track Uncertainty
Target tracking and classification of objects in any sensor data stream plays an important roles in surveillance applications and intelligent gathering. It is however impossible to get consistent and continual track information of the target due to this uncertainty. In this aspect, there have been many studies to develop tracking algorithms that are robust to partial occlusions and that can cope with a short-term loss of observations. This program (working with Sandia National Laboratory) combines the uncertainty in tracking with both spatial and temporal components to develop a comprehensive model that measures and propagated the uncertainties for human analysts to assist in track classification.
Dynamic Mission Planning for Kill Webs of the Future
With advances in Artificial Intelligence (AI), Machine Learning (ML) and Autonomy, Mission Planning for programs in the future need to take embed Human-Machine interactions, information exchange and autonomy into all the all the steps along the way. ARI is working on a program with Defense Advanced Research Projects Agency (DARPA) to build a multi-functional test bed to test these new modalities and algorithms using ground-based and airborne UAV systems in a “capture the flag” game. This will help us achieve the following:
Algorithms for dynamic, decentralized mission planning based on “mission-goals” and Hierarchical Task-Space Specification.
System Portfolio Selection
Learning from Evidentiary Data for Future Operations
Demonstration: Simulation-Environment & UAS test-bed.