The goal is to research and develop a Quality of Information (QOI) Maximizing Information Acquisition Framework (QMAF) for event prediction that goes beyond the limitations of traditional approaches of heuristic/closed form optimization formulations (and its inherent computational complexities) for selection of sensor/data selection. The concept for deriving the QOI optimizing sensor orchestration strategy is looked at as two sub problems a) for a set of random sensor selections, build separate event prediction models and compute the performance accuracies, and b) have a second predictive model learn the relationship between sensor selections and prediction accuracy. The first step calibrates multiple event prediction models for a diverse set of sensor selections, and also generates accuracy data that is used by the second model. The second model can answer the question of the expected accuracy for a given sensor selection, and can point to sensor selection strategies that maximizes the accuracy at minimal acquisition costs. The effort leverages deep learning architectures for complex feature preprocessing, intelligent sampling strategies, as well as active learning algorithms that help supply informative samples for the accuracy prediction problem.
SBIR Phase I
QOI Maximizing Information Acquisition Framework
Agency: Department of Defense
Branch: Air Force
Effective disease-management activities need accurate prediction of disease occurrences. The timely prediction of disease outbreaks facilitates the effective coordination and mobilization of medical, human, and pharmaceutical resources. Prior knowledge of potential disease occurrences enables proactive development of medical interventions, medical prophylaxis to disease hazards, and containment of disease vectors. This is true for both civilian public health coordination and for the military. Infectious diseases are a significant threat to the armed forces and prior knowledge of disease threats will enable the military to not only plan effective mitigation and containment strategies, but also plan the military engagement timing based on these factors. Traditional epidemiology has focused on compartmental models (susceptibility, exposed, infected, recovered (SEIR) based approaches for forecasting disease progression) [Hethcote 2000; Tebbens et. al., 2005; Mayer, 2013]. There is a need to develop occurrence prediction capabilities for novel, emerging, and re-emerging diseases.Continue reading
KBSI applied advanced data fusion technologies in investigating methods for increasing the accuracy of non-destructive inspection imaging technology. Using both ultrasonic and eddy current images of KC-135 lap joint coupons, KBSI developed image processing techniques, both classical and wavelet-based, for the pre-processing of image data prior to corrosion quantification and fusion.
Corrosion is considered the most significant form of damage for an aging aircraft—impacting both maintenance costs and readiness. Currently, there is an urgent need to improve on established non-destructive inspection methods that increase detection reliability and accuracy in multi-layer structures.
GRIPS is a methodology and tool suite that allows military planners to define, systematically evaluate, and globally share data concerning future geo-political contexts. GRIPS’s quantitative analysis and knowledge sharing capabilities help military planners to current and future force deployment plans.
Naval planners must often look years and decades into the future to answer complex questions concerning the objectives and composition of future forces. Successful long-term strategic planning starts with a thorough and systematic evaluation of the future states during which the plans will be enacted, followed by an assessment of plan outcomes and trade-offs. Intelligence Analysts must develop supportable predictions of future geo-political contexts within the regions of interest.
The NODE™ system uses data mining and machine learning technologies to provide a more advanced and adaptable computer network defense. The technology executes data mining and machine learning technologies and algorithms over the network hosts; i.e., over the entire computing fabric.
The ubiquity of computing systems and networks has vastly improved the speed and ease of gathering, storing, and disseminating information. Networking on a global scale, however, also gives rise to a significant disadvantage: network vulnerability. Securing data, particularly with respect to sensitive national security data and data transmissions, is a paramount concern, particularly given the increasing sophistication of computer terrorism.
GERMAT™ provides knowledge-based assistance for neural network and fuzzy logic modeling, helping the U.S. Army to significantly improve battleground classification, target/decoy recognition and discrimination, logistics forecasting, intelligent data fusion, data mining, knowledge discovery, and optimization.
Neural networks and fuzzy set theoretic models offer promising results in performing complex mappings and reasoning in a wide variety of commercial and military domains. Although these technologies have been applied extensively for over a decade, their use in complex, real-time domains is just beginning to be tested. With the advent of recent, more robust and “non black-box”-like algorithms [such as wavenets and Fuzzy Associative Memories (FAMs)], these technologies exhibit even greater promise and potential. These new generation information processing systems provide capabilities like adaptability, robustness, generalization, and the ability to work amid the imprecision and uncertainty of the real world–an ability which makes them especially attractive. Neural networks in particular offer massive parallelism and future promise for hardware implementation and are ideal for applications such as forecasting, classification, pattern recognition, customer analysis, data mining, fraud detection, function fitting, etc.
The MAES initiative designed and developed technology that collects data from multiple sources, analyzes the data for trends and patterns, and, using handheld devices, deliver context sensitive, location based alerts and advice to users on healthcare, disease, safety, and environmental related issues.
Humanitarian relief, disaster recovery, nation building: these missions are becoming more significant for both military and non-governmental organizations every day. Health awareness and a proactive approach to health outcomes are critical to these missions, but, in many developing countries, even basic reporting of the health status of affected populations can be a daunting task. Communications infrastructures are minimal, transportation may be difficult, time is short, and the focus is often on addressing basic human survival needs. Establishing baselines for epidemiological awareness may be too time consuming and difficult to be a priority.
The IIWARS system uses text mining technologies and information fusion to mine data from multiple, distributed text sources, extract features and “indicators” of emerging threats, and improve the DoD’s terrorist threat assessment capabilities.
While analysts and strategists are well versed in tracking threats to conventional military targets, the new asymmetric threats posed by terrorist organizations, as 9-11 has made devastatingly clear, are much more difficult to anticipate. Threats to military targets have traditionally required capabilities that are both expensive and take a long time to develop–activities that satellites and other reconnaissance are more likely to notice.
AARDIS researched advanced electronic warfare (EW) training methods and tools that reduce the cognitive workload of instructors, improve training effectiveness, accelerate improvements in student performance, and reduce training costs.
Pilot training at the U.S. Air Force (USAF) faces two fundamental challenges: rising training costs and restrictions and the changing nature of military conflicts around the globe. According to a recent RAND study, the costs of USAF initial skills training (IST) have risen to approximately $750 million per year. In addition, the opportunities for live training exercises, because of limits on training airspace, rising fuel costs and heightened security, have been diminishing.
KBSI is researching the application of advanced semantic technologies to layered sensing networks (LSN) and the use of data mining technologies for discovering and extracting information from large data stores. The research also investigates the automatic generation of ontologies from unprocessed data sources.
The increased use of networked sensor platforms to perform various command and control, surveillance, and reconnaissance missions has resulted in a corresponding increase in the quantity (and variety) of data that must be organized and analyzed as a precursor to extracting useful information. The purpose of this initiative with Ball Aerospace and Technologies Corporation (BATC) and Air Force Research Laboratories, Anti Tamper and Software Protection (AFRL/RYT) is to research various methods and technologies for extracting valuable information from vast quantities of data.