---------- OSD ----------

94 Phase I Selections from the 12.3 Solicitation

(In Topic Number Order)
Knexus Research Corp.
9120 Beachway Lane
Springfield, VA 22153
Phone:
PI:
Topic#:
(855) 569-7373
Kalyan Moy Gupta
OSD12-AU1      Awarded: 5/23/2013
Title:Decision Support for Anomaly Detection and Recovery for Unmanned System (ADRUS)
Abstract:Unmanned systems have proven their value in combat operations by delivering unprecedented mission performance. Unfortunately, the current unmanned systems are predominantly tele-operated, which tie up many skilled operators per unmanned system. Thus, increasing demands for unmanned system cannot be met with the current state-of-the- art. The problem of low levels of autonomy is further exacerbated by the lack of decision support for behavioral anomaly detection and subsequent recovery planning. We will address this capability gap by developing approaches for increasing the level of autonomy from tele- operated to human supervised as follows. In particular, we will develop ADRUS, a decision support system for anomaly detection and recovery for unmanned system for multi-vehicle missions. ADRUS will provide automated monitoring, perform continuous anomaly detection and analysis in the mission context, analyze root causes for the anomaly and explain its findings to the mission personnel. It will go a step further and recommend plans to recover from the anomaly to minimize disruptions and maximize mission success. To develop these capabilities, we will investigate the use of a variety of probabilistic causal models that exploit the knowledge of mission to assess the deviations and provide accurate alerts. We will investigate fast and incremental automated planning approaches that exploit current resource knowledge to compute effective recovery plans. In developing ADRUS, we will consider human factor issues, such as reduction of cognitive load by developing appropriate alert presentation techniques and human-machine trust by developing decision explanation and justification abilities. We will demonstrate ADRUS feasibility by developing a prototype reference implementation and evaluating it using multi-vehicle mission scenarios.

Pacific Science & Engineering Group, Inc.
9180 Brown Deer Road
San Diego, CA 92121
Phone:
PI:
Topic#:
(858) 535-1661
Maia Cook
OSD12-AU1      Awarded: 5/14/2013
Title:Contextual Anomaly Management Interface (CAMI) for Autonomous System Supervision
Abstract:Unmanned systems are taking on an increasing role in the U.S. military. Monitoring unmanned vehicles, sensors, and events in dynamic environments is an intense and demanding task. Today’s unmanned systems provide limited support for detecting problems and anomalies, and deliver alerts that are generally uninformative and lack context. Given their existing limitations and shortfalls, it is unlikely that today’s technologies and display metaphors will scale to accommodate the increased demands of multi-vehicle and mission management in the future. Building on research results and lessons learned, we propose a novel approach and interface (CAMI: Contextual Anomaly Management Interface) for anomaly management to support effective unmanned systems supervision. PSE has three key elements in place to ensure successful concept development and transition: (1) task and display requirements for anomaly management, (2) an established design process to translate requirements into design, and (3) an established and viable transition plan and customer. CAMI integrates notions from ongoing efforts with innovative concepts for supporting human supervision of automation and anomaly management in a way that respects and balances the strengths and limitations of both the human operator and the inherent capabilities of automation.

Stottler Henke Associates, Inc.
1670 South Amphlett Blvd. Suite 310
San Mateo, CA 94402
Phone:
PI:
Topic#:
(650) 931-2700
James Ong
OSD12-AU1      Awarded: 5/15/2013
Title:Intelligent, Collaborative Management of Autonomous Vehicle Anomalies
Abstract:This project will advance the state of art in two areas necessary for effective exploitation of intelligent diagnosis, planning, and execution technologies within autonomous vehicles. First, it will create and demonstrate the effectiveness of a software framework that specifies how these automation technologies can work together. Second, it will create an intelligent collaborative anomaly management system that enables operators and vehicles to work together as a team, so the benefits of automation can be enjoyed while minimizing the errors that might otherwise result from automation biases.During this Phase 1 project, we will develop a concept of operations and a high-level user interface design; design scenarios describing situations and events that benefit from collaborative anomaly management for autonomous vehicles; design a software framework that specifies the capabilities, algorithms, knowledge requirements, and interfaces of each software module; develop a limited software prototype that illustrates key elements of our approach, and specify metrics for evaluating the utility and efficiency of collaborative anomaly management.

21st Century Systems, Incorporated
6825 Pine Street, Suite 141
Omaha, NE 68106
Phone:
PI:
Topic#:
(808) 748-1825
Amber Fischer
OSD12-AU3      Awarded: 2/15/2013
Title:Autonomously Locating to Perch and Stare (ALPS)
Abstract:Small Unmanned Air Systems (SUAS) are being developed for numerous applications, but size and weight constraints severely limit the capability and endurance of such vehicles. Many missions could be extended by landing for certain periods, entering a low-power state, and re-launching as needed, particularly in urban environments. Technologies have been developed to provide landing gear and flight controls suitable to allow autonomous landing. However, technologies are still needed is the suite of autonomous behaviors necessary to determine when it is appropriate to land, identify a suitable landing zone (LZ), and guide the vehicle to the LZ. While portions of these functions could be performed by a human operator, but it is highly desirable to limit the burden on the operator. 21st Century Systems, Inc. (21CSi) proposes to research and develop ALPS (Autonomously Locating to Perch and Stare), a real-time software solution that autonomously locates potential landing zones from the video feed onboard an SUAS. Based on given operational criteria, such as target surveillance regions and required approach room for landing, ALPS then identifies the optimal site for perching and sufficiently geo-locates the chosen LZ, providing guidance instructions to the flight control software.

American GNC Corporation
888 Easy Street
Simi Valley, CA 93065
Phone:
PI:
Topic#:
(805) 582-0582
Tasso Politopoulos
OSD12-AU3      Awarded: 2/14/2013
Title:Landing Zones Identification (LZI)
Abstract:The objective of this project is to provide the Office of the Secretary of Defense a landing zones determination system for small autonomous air vehicles. The AGNC approach involves the use of EO sensors to effect video data collection and processing for determination of scene constituent elements and objects of interest directly impacting the goal of landing zones determination. The algorithmic aims are structured to accomplish an image perception and understanding framework that results in the synergistic interplay of segmentation and classification algorithms that result in the isolation of the scene constituent elements so that the possible presence of landing zones among them is revealed. The utilized data processing techniques involve both deterministic and probabilistic considerations. Scene segmentation and objects detection algorithms are evaluated as to their effectiveness in accomplishing the automatic realization of the goal of landing zones determination within the context of an overall competing array of other terrain elements and objects in the scene.

Lynntech, Inc.
2501 Earl Rudder Freeway South
College Station, TX 77845
Phone:
PI:
Topic#:
(979) 764-2200
Tony Ragucci
OSD12-AU3      Awarded: 3/5/2013
Title:Landing Site Assessment (LSA) System for Small Unmanned Aerial Systems (SUAS)
Abstract:Perching is an essential tactic for extending the duration of Small Unmanned Aerial System (SUAS) missions. Sustained flight is energetically costly and largely unnecessary during a passive Intelligence, Surveillance, and Reconnaissance (ISR) mission. Data can be collected quietly for extended periods with low-power sensors if the SUAS is positioned strategically. Autonomous landing has been demonstrated in SUAS for well-characterized sites, however unaided identification and evaluation of potential landing zones (LZs) in an unmapped environment remains unsolved and requires considerable abstraction from raw sensor data. Relevant considerations for site evaluation include ability to support the airframe (site slope, roughness, and stability), purview on the area of interest, stealth, ease of access and egress, and suitability for maintaining ground station communications. Lynntech’s Landing Site Assessment (LSA) system processes multiple sensor streams with a multistage layered framework to quickly identify potential landing zones and evaluate them against predefined metrics both autonomously and in real-time. The vision system generates a 3D model of the local environment through varying visibility conditions (day, night, dust, fog, rain, and smoke) and utilizes that data to identify, select, survey, and exploit a locally- optimal landing zone to conduct its mission and return to base.

Scientific Systems Company, Inc
500 West Cummings Park - Ste 3000
Woburn, MA 01801
Phone:
PI:
Topic#:
(781) 933-5355
Ranga Narayanaswami
OSD12-AU3      Awarded: 3/14/2013
Title:ImageNAV-LZ: Autonomous Landing Zone Detection
Abstract:Small Unmanned Autonomous Vehicles (SUAS) are well suited for ISR (Intelligence Surveillance and Reconnaissance) applications. The mission effectiveness of SUAS for ISR can be improved if the SUAS can perch at vantage points, continuing to perform ISR while conserving power. One of the primary bottlenecks to autonomous perching and re-launching is the ability to identify suitable landing zones. The complexity of the urban landscape, combined with the limited processing power on board the SUAS, make this a challenging problem. Scientific Systems Company Inc. (SSCI) in partnership with Brigham Young University (BYU) proposes ImageNav-LZ for autonomous landing zone detection for SUAS. ImageNav-LZ utilizes an innovative spiral search strategy starting just outside the maximum grazing angle of a target surveillance region. As it continues on the spiral, candidate landing-zones are identified when they cast a shadow on the surveillance area. The candidate landing zones are then evaluated for suitability for landing (area, terrain type, slope), suitability for ISR (line of sight visibility of target area) and suitability for re-launch (potential energy, path for take-off, vertical drop etc.). ImageNav-LZ spiral search strategies and efficient vision processing algorithms will significantly reduce the computational burden for landing zone determination.

Near Earth Autonomy
1603 Country Club Dr.
Pittsburgh, PA 15237
Phone:
PI:
Topic#:
(412) 254-3542
Sanjiv Singh
OSD12-AU4      Awarded: 2/15/2013
Title:Cooperative Autonomous Tunnel Mapping
Abstract:While unmanned air vehicles are starting to operate in complex urban and indoor environments, many challenges still remain. Our proposed project will address three key challenges through the development of algorithms. First, we reduce operator workload in the operation of autonomous vehicles in GPS denied environments. Second, the algorithms enable cooperative exploration by multiple and possibly heterogeneous agents. Third, the algorithms will maximize the searchable area in minimal time. The algorithms will be robust so that the team can operate without the need for constant communication. Performance is expected to degrade gracefully when agents spend an increasing amount of time without contact with each other. In Phase I we will develop software simulations to test and validate the cooperative control of multiple agents in performing search through a network of nodes. We will gain significant leverage from existing software in state estimation and mapping that can be used as a base to validate cooperative control. In Phase II we expect to be able to deploy a full team of autonomous flying vehicles performing exploration, mapping, and search. The team consists of members very skilled in the state of the art in estimation, mapping as well as coordination of multi-agents performing search.

Perceptronics Solutions, Inc.
3527 Beverly Glen Blvd.
Sherman Oaks, CA 91423
Phone:
PI:
Topic#:
(703) 342-4684
Timur Chabuk
OSD12-AU4      Awarded: 4/3/2013
Title:Multi-Agent Robust and Scalable Cooperative Indoor Mapping (MARSCIM)
Abstract:The problem of sending a team of robots into an unknown environment to create a map is one of the most studied problems in robotics, and is referred to in the literature as Simultaneous Localization and Mapping (SLAM) or Distributed SLAM (DSLAM) when there are multiple robots. Here we propose to extend existing DSLAM capabilities, with robust and scalable algorithms to support increased cooperation and collaboration between teams of unmanned robotic aerial scouts engaged in tunnel mapping operations. Central to our approach is a set of message passing algorithms for information sharing, frontier expansion, and task allocation that are designed explicitly to address the challenges associated with real world robotics. Specifically, our algorithms are designed to operate under uncertainty and to be highly robust to ongoing communication, robotic, and sensor failures. During mission operations a parameter tuning mechanism, which was trained prior to mission execution using machine learning methods, is used to analyze observed information about the environment, and dynamically adjust algorithm parameter settings in order to maximize performance. In this way, the system continually adapts to match the characteristics of the environment in which it operates. Lastly, we propose a rigorous and thorough simulation- based evaluation of the developed algorithms, in order to prove their efficiency and robustness, and to prepare for transition to real-world robotics.

Physical Sciences Inc.
20 New England Business Center
Andover, MA 01810
Phone:
PI:
Topic#:
(978) 689-0003
David R. Manegold
OSD12-AU4      Awarded: 2/26/2013
Title:MATCH-SLAM: A Multi Agent Tasking and Control metHod for Simultaneous Localization and Mapping
Abstract:Physical Sciences Inc. (PSI) proposes MATCH-SLAM, an innovative “Multiple Agent Tasking and Control metHod for Simultaneous Localization and Mapping” for multi-agent cooperative and autonomous exploration of tunnels, caves, indoor, or other GPS denied environments. Current tunnel mapping efforts using autonomous ground vehicles are limited by platform mobility and power requirements, while aerial approaches currently lack robustness, effective coordination methods, and autonomy. PSI’s design gives specific attention to the complex interactions between heterogeneous agents and how to fuse their disparate sensor types, the constraints of limited on-board computational resources, limited platform mobility, and bandwidth-limited communications. This approach promises to enable efficiently coordinated exploration and mapping by many classes of autonomous systems, including PSI’s own highly capable, highly mobile, low-power, low-cost micro-air vehicle. This is achieved using PSI’s novel planning algorithms which make use of a multi- dimensional cost/score matrix for heterogeneous agent cooperative tasking, specific agent actions behavior patterns tailored to reduce crucial uncertainties, and a state estimator designed to maintain map consistency across agents in the presence of constrained or intermittent communication. PSI proposes to test and demo this technology during the program both in software simulations and on its own micro-air vehicle platforms in representative environments.

Scientific Systems Company, Inc
500 West Cummings Park - Ste 3000
Woburn, MA 01801
Phone:
PI:
Topic#:
(781) 933-5355
Mads Schmidt
OSD12-AU4      Awarded: 3/12/2013
Title:AIRTEAM: Aerial Indoor Robot Teams for Exploration And Mapping
Abstract:For the AIRTEAM (Aerial Indoor Robot Teams for Exploration And Mapping) Project, Scientific Systems Company, Inc. (SSCI) and iRobot Corporation propose to develop a command-and-control algorithm to allow aerial robotics scouts to cooperatively explore an unknown indoor environment and communicate their findings to each other and their human operators. In Phase I of this project, we will study the feasibility of cooperatively searching indoors with aerial robots. We will develop a basic multirobot coordination and exploration planning algorithm, and we will test this algorithm in simulation. This algorithm will be capable of handling the challenges of indoor operations, such as intermittent wireless communications and obstacles.

Aptima, Inc.
12 Gill Street Suite 1400
Woburn, MA 01801
Phone:
PI:
Topic#:
(781) 935-3966
Stacy Pfautz
OSD12-AU5      Awarded: 2/28/2013
Title:AWAKE: Adaptive Workspace for Analyst Knowledge & Engagement
Abstract:Increasingly there is the need to process large volumes of disparate data containing critical information that needs to be analyzed in a timely, efficient matter. While tools exist to support information search, retrieval and exploitation, these tools do not adequately support the analyst since they lack understanding of the semantics of the information, how the information relates to the analytic task, and how the analyst fuses the information in assessing a situation. Cognitive limitations to analysts’ attention and working memory expose them to cognitive biases, and current tools do not provide any help in guarding against analytic vulnerabilities. Aptima and our partners, propose to develop an Adaptive Workspace for Analyst Knowledge and Engagement (AWAKE). AWAKE represents the next generation of cognitive, knowledge-aided analyst support systems that promotes a more effective human- machine partnership so that analysts can focus on what they uniquely do best as humans, and the autonomous system “looks over their shoulder” to provide them cognitive aid. AWAKE will provide a capability for measuring and assessing the analyst’s level of rigor; automatically identify indicators of cognitive biases and vulnerabilities, based on a semantic interpretation of the user’s interactions with the system.; and personalize agents to actively support analysts’ activities.

Charles River Analytics Inc.
625 Mount Auburn Street
Cambridge, MA 02138
Phone:
PI:
Topic#:
(617) 491-3474
Sean Guarino
OSD12-AU5      Awarded: 3/1/2013
Title:Services for an Individualized Dynamic Environment using Knowledge, Information, and Context for Workspace Management (SIDEKICK)
Abstract:Intelligence analysts are flooded with complex data, creating failures and delays in exploiting critical information. Attempts at automated decision-aids have failed because they ineffectively strive to supplant the analyst, rather than enable collaboration. Instead, analysts manually sift through data, leading to human errors from skill and knowledge gaps and human biases (e.g., confirmation bias). Analysts need automated assistants that provide a collaborative human-automation partnership, forming a common ground with the analyst that drives workspace management for information observability. To address these needs, we propose to design and demonstrate Services for an Individualized Dynamic Environment using Knowledge, Information, and Context for Workspace Management (SIDEKICK). SIDEKICK builds on our team’s previous research designing a framework for intelligence analysis, expanding on iterative broadening and narrowing with a Joint Activity Theory-based approach for automation-analyst coordination. SIDEKICK includes: (1) an Analysis Modeling Service that uses Probabilistic Relational Models to provide a robust modeling framework that can produce a common ground representation of the analysis based on observations of analyst behaviors; and (2) a Workspace Adaptation Service that uses case-based reasoning to analyze the common ground representation to recommend selections and adaptations of workspace tools and visualizations.

Securboration Inc
1050 W NASA Blvd Suite 155
Melbourne, FL 32901
Phone:
PI:
Topic#:
(919) 217-7269
Dr. Bridget McInnes
OSD12-AU5      Awarded: 2/27/2013
Title:Fashioning of an Adaptive Workspace through Autonomous Services
Abstract:The Intelligence Community (IC) has applied significant effort towards creating automated adaptive workspaces. One success in this area originated with DARPA’s Personalized Assistant that Learns (PAL), which focuses on improving the way that computers support humans through the use of cognitive systems that reason, learn from experience, and accept guidance in order to provide effective, personalized assistance. While PAL continues to achieve success in IC automation technologies, there are other avenues for future work which focus on a more bi-directional human-IT partnership. What analysts really need is an adaptive collaboration between the human and machine – one in which the computer learns, adapts, and informs analysts based on their practices, situational context, and data content. This capability is analogous to how paralegals function as intelligent, dynamic, and indispensable partners for lawyers in the legal community. To provide this next generation capability, Securboration is teaming with SRI International and Intelligent Software Solutions (ISS) to develop the Adaptive Data Immersion ENvironment (ADIEN). Our approach for ADIEN leverages Securboration’s expertise in machine learning and reasoning; SRI’s leadership on the PAL program and adaptive, personalized intelligent assistance technologies; and ISS’ subject matter expertise and leadership in developing and deploying operational intelligence systems to the IC.

Stottler Henke Associates, Inc.
1670 South Amphlett Blvd. Suite 310
San Mateo, CA 94402
Phone:
PI:
Topic#:
(617) 902-2223
Eric Domeshek
OSD12-AU5      Awarded: 2/26/2013
Title:A Monitoring and Adapting Intelligence Assistant (MAIA)
Abstract:Intelligence analysts need help managing a barrage of tasks and flood of information while overcoming cognitive limitations. That help must respond to their individual time-varying needs and context. Challenges include (1) sensing analysts’ situations; (2) tracking their interests and needs; (3) managing information flows; (4) facilitating user focus and understanding; and (5) controlling and orchestrating the above capabilities. The keys to adaptive analyst support include (a) access to user/situational data; (b) characterization of the relevant features latent in that data; (c) correlating features with known cognitive biases, failure modes, and support needs; and (d) application of appropriate learning and control mechanisms to map from situational features to system actions. We propose to develop technology for a Monitoring and Adapting Intelligence Assistant (MAIA). During Phase I we will identify use cases and requirements; use those to flesh out the general challenges and capabilities sketched above; outline a technology and follow-on research framework to address those needs; and design and prototype subsets of the overall vision to demonstrate feasibility. We expect to focus development work on mapping from user activity to metacognitive prompts and information-sharing actions. Finally, we will develop a Phase II plan to follow through on design, development, and transition.

Aptima, Inc.
12 Gill Street Suite 1400
Woburn, MA 01801
Phone:
PI:
Topic#:
(781) 496-2467
Georgiy Levchuk
OSD12-AU6      Awarded: 2/25/2013
Title:Sense-making via Collaborative Agents and Activity Networks (SCAAN)
Abstract:The quantity of data that need to be collected, examined and shared during Intelligence, Surveillance, and Reconnaissance Operations is growing fast due to increasing use of sensors. To deal with this challenge, the Air Force intelligence services are implementing a new process of planning and direction, collection, processing and exploitation, analysis and production, and dissemination (PCPAD). PCPAD requires new technologies that collect and process only the most critical and relevant information. Aptima proposes to develop a system for Sense-making via Collaborative Agents and Attributed Networks (SCAAN) that integrates distributed situation understanding, autonomous knowledge seeking, dynamic collaboration, and an adaptive heterogeneous command and control organization. SCAAN will solve challenges of collaborative large-scale information processing by incorporating a model of dependencies between essential elements of information based on real-world processes into its distributed information sharing framework. These dependencies will be used to construct an agent organization, which assigns command and execution roles to sensor nodes and is required for reducing complexity of managing heterogeneous sensor team, and a collaboration policy, which will be based on dependencies between the tasks executed by different nodes. SCAAN will achieve reduction in data analysis time while maintaining optimality of situation estimates obtained in a distributed manner.

EDAptive Computing, Inc.
1245 Lyons Road Building G
Dayton, OH 45458
Phone:
PI:
Topic#:
(937) 281-0792
Adam Langdon
OSD12-AU6      Awarded: 2/26/2013
Title:Autonomy for Seeking, Understanding, and Presenting Information
Abstract:Modern net-centric warfare produces and consumes a tremendous amount of data. This data is produced by multiple systems over time and is in different forms and formats. This can easily become overwhelming to an analyst, who must be able to quickly determine the impact of anomalous events or changing trends. New methods are needed to further filter data and extract more meaningful relationships among facts discovered across multiple datasets. The success of the warfighter is not only threatened by the sheer volume of information, but by the increasing risk of conflicting and misleading intelligence. This problem must be addressed for the DoD to remain successful on an increasingly chaotic battlefield. In response to these issues, Edaptive Computing, Inc. presents an innovative solution that specifically addresses the requirements of the OSD and provides a significant step toward analyzing multiple, large, and heterogeneous datasets. This approach consists of data mining agents employing various advanced techniques implemented on scalable distributed platforms. As a result, these techniques can discover new relationships between current events, other intelligence, and historical data. Ultimately, the GUARDS solution allows an analyst to perform critical decision-making with confidence and clear awareness of the situation.

InferLink Corporation
2361 Rosecrans Avenue Suite 348
El Segundo, CA 90245
Phone:
PI:
Topic#:
(310) 383-9234
Steven Minton
OSD12-AU6      Awarded: 2/25/2013
Title:Autonomy for Seeking, Understanding, and Presenting Information
Abstract:There is an enormous amount of amount of data available on the Web. The richness of this data has made it a priority to develop intelligent systems that can mine and make sense of data in order to assist analysis in predicting and explaining events around the world. In this project, we propose to develop a scalable system to intelligently assist analysts by autonomously collecting, analyzing and presenting information. To address the challenge of determining what data is worthwhile for the system to collect, and what is most valuable for the analyst to see, we propose an architecture based on decision-theoretic search control. In this paradigm, the system estimates the utility of alternative options, enabling it to make rational choices, when faced with uncertainty, and a complex set of alternatives. Moreover, machine learning algorithms make it possible to estimate the utility of different alternatives based on experience. A second challenge we address is how to organize a complex system that can collect, interpret and mine large amounts of heterogeneous, multimodal data. We propose to employ a joint inference approach that allow components to flexibly communicate their findings, and in effect, work together to produce optimal results.

ADA Technologies, Inc.
8100 Shaffer Parkway Suite #130
Littleton, CO 80127
Phone:
PI:
Topic#:
(303) 792-5615
Thierry Carriere
OSD12-EP3      Awarded: 4/26/2013
Title:Ultra-Safe Enclosure for High Energy Density Devices: Thin Laminate Walls and Flameless Venting
Abstract:The weight, volume and storage capacity benefits of current and future high energy density storage systems make them attractive options to complete far-reaching and ambitious missions. Safe packaging of these devices has lagged and as a result, a few significant events have taken place, highlighting the need for better tools to manage and control the uncontrolled release of stored energy. ADA Technologies is proposing to develop and test a new, rugged, ultra-safe enclosure meeting the DoD requirements by leveraging our deep technical expertise in materials and fire safety. The selected approach is to combine the properties of several low-cost, lightweight and off-the-shelf materials in a laminate structure which will be able to channel and contain any released energy and protect nearby equipment and personnel. Further improving the safety of the enclosure and the surroundings is the inclusion of a flameless vent. ADA Technologies has established several key collaborations to enable successful technical development and future commercialization of the technology in both military and commercial markets. Our partners include the technical experts at the NASA Johnson Space Center, DoD prime contractor Lockheed-Martin, commercial companies A123 Systems (lithium ion batteries), Newtex (fire-resistant materials) and Hoffman (industrial enclosures).

Paragon Space Development Corporation
3481 E. Michigan Street
Tucson, AZ 85714
Phone:
PI:
Topic#:
(520) 382-4812
Grant Anderson
OSD12-EP3      Awarded: 6/26/2013
Title:Coacting High Integrity Materials Energetic Release Attenuation (CHIMERA)
Abstract:Paragon Space Development Corporation (Paragon) and our subcontractor, Thin Red Line Aerospace (TRLA) will create a Coacting High Integrity Materials Energetic Release Attenuation (CHIMERA) system that will meet the volume, thermal, ballistic, pressure, electro-magnetic (EM) and chemical containment requirements while providing modularity and significant weight savings over existing enclosure options. The CHIMERA system will build upon the Paragon/TRLA team’s extensive experience developing, modeling, simulation, and test of inflatable space structures able to withstand the rigors of that environment, including impact from micro-meteorites and orbital debris (MMOD) to create a system that will enable a safe operating environment for personnel and sensitive equipment. Expert selection of advanced ballistic fabrics, design of rigid skeletal structure, and integration of novel phase change material (PCM) thermal mitigation methodologies will provide the customer with an adaptable, scalable, modular solution that is capable of reacting in a deterministic manner to any credible failure mode of high power energy storage devices including but not limited to batteries, ultra-capacitors, and flywheels. CHIMERA will be modular, significant reducing development or installation issues for new applications and will include optional doors and view ports, and be relatively light weight, highly transportable and can be installed on-site to protect existing infrastructure.

Scimitar Technologies LLC
2005 Big Horn Drive
Austin, TX 78734
Phone:
PI:
Topic#:
(512) 692-9663
Brian Muskopf
OSD12-EP3      Awarded: 5/1/2013
Title:Energy Storage Enclosure Technologies for High Density Devices
Abstract:Energy storage systems comprised of high-density batteries, capacitors and/or flywheel technologies provide numerous benefits for shipboard power systems. However, high-density energy storage systems which must be installed in locations populated by personnel and sensitive equipment can fail catastrophically. Robust enclosure designs and materials are required that are capable of overcoming effects related to temperature, pressure and inertial effects at the same time. Innovative enclosure designs are required because the size and weight of the enclosure cannot substantially reduce the power density of the energy storage system. The enclosure cannot expand the volume of the racked storage components by more than 10%. This project will develop a cost effective, lightweight, corrosion and fire resistant, robust composite Energy Storage System Protective Enclosure that will contain all electrical, fire, chemical and inertial hazards of the energy storage system while meeting the structural and environmental exposure design requirements for use onboard U.S. Navy ships. The protective enclosure will be constructed from composite materials that will meet all U.S. Navy design requirements including fire, smoke and toxicity, and all environmental and chemical exposures. The composite protective enclosure design will not expand the volume of the racked storage components by more than 10%.

Management Sciences, Inc.
6022 Constitution Ave NE
Albuquerque, NM 87110
Phone:
PI:
Topic#:
(505) 255-8611
Kenneth D. Blemel
OSD12-EP4      Awarded: 7/26/2013
Title:Embedded Collaborative Intelligent Generator Management with Grid-Centric Microcontrollers
Abstract:In Afghanistan, fuel to run electrical generators is a major cost driver. The DoD’s latest method is “right size” with units with multiple generators of different ratings and train soldiers to manually switch generators as demand changes. Typically, demand has wide swings throughout the day and soldiers can’t just stand next to generators to “right size.” Inevitably, fuel will be wasted waiting for soldiers to throw the switch. MSI has developed an intelligent control module that employs machine learning to increase energy efficiency of a single tactical generator. Our innovation is to create a new wireless automated switch module that networks with other modules on the grid. They will use alternative energy first and bring the right generators online when alternative and stored energy is insufficient. When demand drops the modules will reverse the strategy to continuously maximize fuel savings 24x7. Our product vision is an install kit packaged to replace the existing manual switch boxes so that retrofit will be quick and easy. Phase I will develop an architecture and a small form factor prototype and demonstrate right sizing three generators and alternative energy sources in realistic conditions.

OPEX SYSTEMS LLC
3306 Greencastle Chase NE
Marietta, GA 30062
Phone:
PI:
Topic#:
(404) 268-5522
Kojakully (Ram) Shetty
OSD12-EP4      Awarded: 8/5/2013
Title:A Standards Based, Open-Architecture, Ruggedized Tactical Microgrid Controller
Abstract:The Army is addressing a technical challenge of constrained data interface to vendor supplied equipment, to integrate Energy Storage (ES) and tactical Energy Assets (EA) to improve fuel efficiency and reduce wear. The Phase I effort will investigate the feasibility of developing a ruggedized Intelligent Power Management Controller (IPMC) which will act as (a) an Universal Data Integration Gateway to any vendor supplied EA/ES, and (b) an Automation Supervisory Controller to achieve the Army’s objectives in tactical Microgrid configurations. The firm is proposing to leverage their extensive experience in advanced industrial and electric grid automation technologies for this project. Three innovations are proposed: a) industrial best practices applied to the military application, b) Standards based communication protocols for data integration, and c) a modern digital control scheme. Anticipated benefits include lower product cost, ease of use, open architecture, standardization and low learning curve. Proposing team includes personnel from Battelle, top non-profit R&D organization with extensive experience in developing solutions for national security applications. The team has capabilities to develop a practical and innovative IPMC solution that will reduce the technology risk and time-to-deployment. The firm estimates excellent dual-use market potential and intends to commercialize IPMC through its energy technology business.

PC Krause and Associates, Inc.
3000 Kent Avenue, Suite C1-100
West Lafayette, IN 47906
Phone:
PI:
Topic#:
(765) 464-8997
Marco Amrhein
OSD12-EP4      Awarded: 7/24/2013
Title:Intelligent Power Management Module for Autonomous Power Generator Operation
Abstract:The US Army and Marines rely significantly on fossil-fuel based tactical power generation to provide electrical energy to deployed troops in small camps and forward operation bases. These generators are sized to support the expected maximum power but are often operated far below rated power due to the dynamic nature of electrical loads. On average, the generators are loaded less than 50% of the available peak power capability. Inherently, these generators have low energy conversion efficiencies when operated significantly below rated power, which equates to wasted fuel in order to provide power capacity that is unused during the majority of operation. Thus, technology innovations are needed that allow for better utilization of a generator's power capacity when said generator is online, while at the same time providing the on-demand dynamic load power desired by the attached equipment.PC Krause and Associates (PCKA) has previously demonstrated a power buffer concept that addresses mismatches between dynamic loads and generation capabilities on modern more- electric aircraft. The patent-pending power-sharing control concept is, with minor modifications, applicable to the challenges addressed by this proposal. PCKA proposes an intelligent power-management module (IPMM), consisting of an energy storage device, power electronic converters, and appropriate power-sharing controls, to act as a buffer between a tactical generator and dynamic loads. The IPMM will essentially function as an energy reservoir; charging and discharging its energy storage device as appropriate to ensure that load demands are met while driving the operation of the tactical generator into a more efficient operating regime. Due to this new loading profile seen by the generator, fuel conversion efficiency will significantly increase (estimated improvement of 25%), as the generator is either operated at full capacity or turned off. Note that the IPMM will be autonomous and non-invasive, i.e., it will not require any adjustment in existing generator controls or hardware, a desirable benefit considering the target application.The primary objective of this Phase I proposal is to develop the proof-of-concept IPMM, including the design of a hardware architecture and power-sharing control strategies. To demonstrate the feasibility of the proposed IPMM design and justify the transition from Phase I to Phase II, a combination of advanced modeling, simulation, and analysis and hardware prototyping will be utilized to demonstrate that the proposed IPMM is capable of meeting the relevant performance specifications. If successful, the IPMM technology will be demonstrated as a full-scale prototype in Phase II and ultimately transition to DoD programs under a Phase III contract.

Radiance Technologies Inc.
350 Wynn Drive
Huntsville, AL 35805
Phone:
PI:
Topic#:
(334) 246-2871
Brian Freeman
OSD12-EP4      Awarded: 5/30/2013
Title:Tactical Power Plant Multi-Generator Intelligent Power Management Controller
Abstract:"Diesel generators provide the majority of power generation in forward operating areas, frequently operating inefficiently and requiring excessive maintenance. “Right size” generator operation, energy storage, renewable energy inputs, and intelligent power management strategies can greatly reduce fuel consumption and reduce the level and frequency of maintenance. Predictive load profiling strategies, integration, and communications between the multiple devices on a micro-grid will yield additional fuel savings.Radiance Technologies, Inc. proposes to design and build a highly configurable power manager to optimize generator operation, reduce fuel consumption, and seamlessly integrate with multiple devices with and without external communications. The power manager consists of an energy storage device, inverter/charger, generator-synchronizer, renewable energy input, and intelligent controller. The proposed system will operate as a 10 kW redundant power plant or a 20 kW power plant using the new AMMPS generators or the legacy TQG generators; accepting 10 kW inputs from AC sources and one 3 kW renewable source such as solar. All configurations are packaged onto a Light Tactical Trailer (LTT) with two 10 kW generators.Radiance will establish an open communications protocol and implement drivers to interface with existing and future external communications protocols."

Global Technology Connection, Inc.
2839 Paces Ferry Road Suite 1160
Atlanta, GA 30339
Phone:
PI:
Topic#:
(770) 803-3001
Freeman Rufus
OSD12-EP5      Awarded: 2/1/2013
Title:Dynamic Time and Frequency Domain Modeling of Aircraft Power System with Electrical Accumulator Units (EAU)
Abstract:Global Technology Connection, Inc. (GTC) seeks to develop generic time and frequency domain analysis modeling and analysis tools to analyze and determine mitigation strategies to maintain power quality with high dynamic aircraft power systems operating with electrical accumulator units (EAU). Phase I effort will concentrate on the initial requirements and design of a mixed / multi-level modeling concept with the logic to select an appropriate level of modeling of the interacting subsystems that would allow capturing the essential features of the phenomenon under investigation that a user wants to study (system stability, power system response to faults, power system quality for loading and regeneration, etc.). Linearized, behavioral and detailed models of aircraft power subsystems will be implemented in Simulink. We will determine mitigation components and filtration strategies for generic EAU able to source or sink 150kW for 100ms or 50ms regenerated energy, respectively. The modeling and analysis tool will be demonstrated using a power distribution system of a transport aircraft utilizing EAU, energy storage and mitigation components. The modeling tool’s performance will be assessed according to speed, accuracy and the capability of the designed mitigation components with filtration strategies in adhering to MIL-STD-704F. Phase I will include the development of plans to further increase fidelity of modeling tool to incorporate degradation in the energy storage systems. Phase II will develop and refine the fidelity of the Phase I modeling and analysis tool and include the development of improved mitigation components and methodologies that are validated in a simulated power system bus incorporating signals for high demand loads, sources, and energy storage.

NextWatt LLC
1635 Westbury Drive
Hoffman Estates, IL 60192
Phone:
PI:
Topic#:
(315) 543-9085
Sudip K. Mazumder
OSD12-EP5      Awarded: 1/31/2013
Title:Multi-Scale Hybrid Modeling Based Fast Component/System Reliability Assessment Analytical Tool for EAU Buffered Aircraft Power System
Abstract:Technical Objective 1: Develop a scalable and plug-and-play unified hybrid multi-scale, multi-resolution, and multi-domain modeling framework for component and system level representation of interactive aircraft power system comprising varied energy generation sources, EAUs, complex loads and load patterns, and power distribution and control system. Technical Objective 2: Develop advanced analysis methodologies for investigating evolving behavioral dynamics, stability, and reliability of a complex aircraft power system with changing operating conditions, changing component and system dynamics, and changing power-system network topology and connectivity. Such an analysis methodology based on advanced nonlinear techniques (for large-signal/large-perturbation conditions) as well as linear techniques for small-signal analysis will preclude the need for rapid, robust, and in- depth time and frequency system analyses without resorting to the conventional approach of computationally intensive large number of simulations which is often inadequate to capture nonlinear system effects. Technical Objective 3: Develop a software platform for capturing component and system models and for capturing algorithmic tools for robust component and system reliability and interaction analysis to demonstrate the efficacy of the tools developed under Objectives 1 and 2 and conduct comprehensive scenario centric case studies leading to pathway for optimal system and component designs.

PC Krause and Associates, Inc.
3000 Kent Avenue, Suite C1-100
West Lafayette, IN 47906
Phone:
PI:
Topic#:
(765) 464-8997
Jason Wells
OSD12-EP5      Awarded: 2/19/2013
Title:Dynamic Time and Frequency Domain Modeling of Aircraft Power System with Electrical Accumulator Units (EAU)
Abstract:The movement to more-electric architectures during the past decade in military and commercial airborne systems continues to increase the complexity of designing and specifying the electrical power system. The addition of numerous high-power electric loads has drastically altered the dynamics of power flow on the electrical bus. Many of these loads often exhibit peak-to-average power ratios in excess of 5-to-1 for brief periods of time (50- 5000 ms). In addition to this high peak-power, some of the loads can produce regenerative power flow equal to their peak power draw for brief periods of time (typically 20-200 ms). Such load characteristics, coupled with complex and varied source characteristics under varying electrical power system configurations, can result in undesirable system performance from both a dynamic-transient and a spectral-content (frequency-domain) perspective. In order to mitigate such undesirable performance, modern electrical power system designers must develop and apply suitable mitigation strategies, which typically involve energy storage, filtration, and/or advanced control. Since weight and volume are significant concerns on airborne platforms, it is necessary to quantify whether the proposed solution’s impact on power quality is justified in consideration of the impact on weight and volume to the platform; however, limited tools exist for performing such analysis at the integrated-system level from the dynamic-transient and frequency-domain perspective. As such, the primary objective of the proposed SBIR program is to develop such time- and frequency-domain analysis models and tools to enable the identification, development, and analysis of emerging and future mitigation strategies that can maintain power quality in the presence of ever increasing dynamic load requirements. It is anticipated that such tools will be sufficiently generic that they can be applied broadly to a wide class of platforms and technologies yet sufficiently customizable to a target application such that they can provide meaningful insight into specific technology development decisions. In the Phase I effort, PC Krause and Associates will develop and demonstrate such tools at the proof-of-concept level. The developed tools shall be capable of analyzing the impact of emerging and future components, control strategies, and architecture(s) on electrical power quality through the prediction of key performance aspects of both the technology under investigation and the remainder of the electrical power system into which the technology is being inserted. In the Phase I effort, a proof-of-concept demonstration shall be performed by analyzing the impact of inserting an electrical accumulator unit capable of sourcing a minimum of 150 kW for 100 ms and sinking a minimum of 150 kW for 50 ms. At a minimum, the impact analysis and performance predictions generated by the tool shall be capable of determining the adherence of the technology under investigation, and the resulting electrical power system, to MIL- STD-704F, which is a requirement for the successful transition of most technologies that might be investigated for utilization in current and future generations of airborne platforms.

IntraMicron, Inc.
368 Industry Drive
Auburn, AL 36832
Phone:
PI:
Topic#:
(334) 502-2973
Hongyun Yang
OSD12-EP6      Awarded: 4/17/2013
Title:Advanced thermal management materials and strategies for packaging high rate cylindrical cells and battery packs
Abstract:Microfibrous media (MFM) composed of sintered micron diameter copper fibers will be evaluated for its ability to transfer heat from cylindrical cell arrays (batteries and capacitors) to cooling coil and heat transfer surfaces. The void volume of the MFM will be from 70 to 90% and the resulting porosity filled with a suitable phase change material (PCM) as an added (and intimately contacted) thermal buffer. Preliminary calculations and data indicate that a 30 vol.% MFM filled with 70 vol% PCM will have a thermal conductivity of 60 W/m-K, an enthalpy capacity between 25 and 60C of 249 KJ/liter, and due to its compliant nature an interfacial heat transfer coefficient of 420 W/m2-K. Compared to other documented technologies this combination of properties is superior for both brief periods of high rate discharge in a near adiabatic fashion (with rapid thermal recovery) and also extended duration applications at high average power and thermal dissipation levels. The porous media is cost effectively manufactured on high speed paper machines and readily packaged between battery arrays and heat transfer surfaces. The approach is suitably robust for military applications, maintains appropriate cell temperatures, is mass and volume efficient, and can be easily packaged to avoid shorts.

Thermal Management Technologies
1750 North Research Park Way Suite 104
North Logan, UT 84341
Phone:
PI:
Topic#:
(435) 755-6401
J. Clair Batty
OSD12-EP6      Awarded: 6/6/2013
Title:Two-Phase Immersion Cooling for Cylindrical Geometry Energy Storage Components
Abstract:Thermal Management Technologies proposes to develop concept designs for a scalable two-phase liquid-vapor immersion cooling system and perform a subscale proof of concept demonstration of the technology. Direct immersion cooling is capable of providing thermal management of individual cells whether in the cylinder, prismatic, or pouch configuration and maintaining a very nearly uniform temperature distribution among all cells. The proposed design, which employs a high quality dielectric working fluid in direct contact with the energy storage components, will be shown to provide effective thermal management of cylindrical components without decreasing their packing density. Overcoming the deficiencies of existing cooling approaches, the proposed system provides efficient thermal coupling between energy storage components and a cold sink by utilizing the very high boiling and condensation heat transfer coefficients of the working fluid. This eliminates the inefficiencies of conduction through interface materials or electrical connection terminals. An automatic degas module is proposed to simplify assembly, fluid filling, and maintenance procedures and maintain the working fluid in a saturated state.

ThermAvant Technologies, LLC
1000 A Pannell Street
Columbia, MO 65201
Phone:
PI:
Topic#:
(415) 264-0668
Joe Boswell
OSD12-EP6      Awarded: 6/6/2013
Title:Cylindrical Geometry Energy Storage Cooling Architectures
Abstract:ThermAvant Technologies, LLC and University of Texas at Arlington propose to design, model, fabricate and empirically test conformal fitting cylindrical energy storage heat cooling packages. ThermAvant will optimize cylindrical battery array cooling solutions that balance size, weight, coolant flow rate and pressure drops, maximum temperatures, thermal responsiveness, and temperature uniformity. Two different approaches will be evaluated: micro-channel single-phase cold plates; and cold plates with two-phase super-thermally conductive oscillating heat pipes (OHPs). A number of innovative manufacturing techniques will be evaluated and compared to conventional micro-channel forming and sealing processes. University of Texas at Arlington will provide the high fidelity electrical and thermal test facilities to evaluate thermal solutions' effectiveness in cooling real-world cylindrical battery cells.

TIAX LLC
35 Hartwell Avenue
Lexington, MA 02421
Phone:
PI:
Topic#:
(781) 879-1269
John Dieckmann
OSD12-EP6      Awarded: 4/26/2013
Title:Cylindrical Geometry Energy Storage Cooling Architectures
Abstract:Electrical energy storage devices – e.g., rechargeable batteries, ultracapacitors, flywheels – all suffer some level of energy conversion loss as they are charged and discharged. Lithium-ion and other lithium-based rechargeable batteries lose electrical energy to IR voltage loss as current flows through the anode and cathode electrodes. Flywheel systems similarly dissipate heat through loss of electrical energy. In energy storage systems intended to operate at high power levels, with full charge and discharge cycles on the order of minutes in length, active cooling is required to remove the dissipated heat and limit the temperature of the battery or flywheel within limits consistent with maintaining performance and long cycle life. TIAX proposes a scalable cooling system architecture that has an effective thermal connection to each cylindrical energy storage device and can connect to a cooling source, for example, a fresh water cooling loop, that would be accessed at the back plane of the cabinet holding the battery racks. Preliminary thermal modeling shows that this cooling architecture will provide the necessary cooling performance using a fresh water cooling loop that is cooled via heat exchange with sea water, with a resulting fresh cooling water supply temperature as high as 40 degrees C.

5-D Systems Inc.
1 Chisholm Trail, Suite 3200
Round Rock, TX 78681
Phone:
PI:
Topic#:
(512) 238-9840
Bennie Ray Kirk, Jr.
OSD12-EP7      Awarded: 6/3/2013
Title:Militarized Power Line Communication
Abstract:5-D is developing a cost effective, militarized, power line communications (PLC) system leveraging commercial “Smart Grid” and PLC technologies. Our approach provides the system components necessary to properly manage the variable resources and loads inherent in micro grid power systems like those employed at forward operating bases. In addition, our solution offers the option of providing broadband networking communications, further minimizing the infrastructure and installation typically required. The system components under development are modular by design, offering flexibility of integration and application for various power grid issues like seamless integration of renewable power sources and generator load balancing. 5-D Systems has already designed and proven militarized PLC technology for the Air Force Research Lab (targeted for legacy aircraft and other applications) and will leverage this experience to design, demonstrate, and ultimately deliver a viable micro grid communications and control system.

Physical Optics Corporation
Information Technologies Division 1845 W. 205th Street
Torrance, CA 90501
Phone:
PI:
Topic#:
(310) 320-3088
Alireza Shapoury
OSD12-EP7      Awarded: 6/3/2013
Title:Communicator for Electrical Grid Axillary Services
Abstract:To address the OSD’s need for a militarized power line communication system, POC proposes to develop a new Communicator for Electrical Grid Axillary Services (CEGAS). CEGAS is based on a set of innovative modifications to the available power line communication systems that will enable the OSD to enhance resource management securely and reliably without the need to install and maintain a parallel data network for grid auxiliary services such as peak shaving, frequency control and contingency protection. As a result, this system offers militarized power line communication for control and management of fuel and several other renewable energy sources in a tactical electrical grid. In Phase I, POC will demonstrate the feasibility of CEGAS by analysis and design of a communication protocol and fabrication of a prototype to prove key performance areas, such as link availability, interoperability, interference robustness, latency, and security for a 208 VAC, three-phase electrical grid. We will further provide a cost/benefit analysis and a development plan with performance goals for Phase II. In Phase II, POC plans to develop a full system and test it with the OSD; perform a detailed cost analysis of CEGAS compared to existing technologies; and provide a Phase III development plan.

American GNC Corporation
888 Easy Street
Simi Valley, CA 93065
Phone:
PI:
Topic#:
(805) 582-0582
Tasso Politopoulos
OSD12-ER1      Awarded: 4/4/2013
Title:Graphical Evolutionary Hybrid Neuro-Observer (GNeuroObs) System
Abstract:The core project objective is to demonstrate the Graphical Evolutionary Hybrid Neuro- Observer (GNueroObs) System in conducting complex system workflow analysis for visualizing the interaction between components and subsystems. This software tool provides a state-of-the-art systems analysis framework compiling technologies involving: (a) advanced system and degradation modeling techniques using hybrid schemes with dynamic event network modeling; (b) novel automated knowledge generation; (c) high performance health monitoring (fault diagnosis) methodologies; and (d) advanced object-oriented software engineering practices. A novel methodology is developed and tailored to evaluate faults and engineering changes and their propagated effects within complex systems. This is made possible by understanding intrinsic natural behavior as well as degradation tendencies for entities (subsystems and components). The software toolset satisfies current OSD needs by including the capabilities of: (i) system decomposition; (ii) modeling of entity interrelations; (iii) graphical representation; and (iv) quantitative information overlay. The software is verified and validated with a representative target system. The overall expectation is that this software will enable engineers to obtain a more realistic understanding of complex system behavior.

CyDesign Labs, Inc.
1810A Embarcadero Road
Palo Alto, CA 94303
Phone:
PI:
Topic#:
(760) 310-5061
Michael Koopmans
OSD12-ER1      Awarded: 6/18/2013
Title:Evaluating Component Interactions Within Complex Systems
Abstract:Understanding complex system behavior and designing resilient systems is becoming progressively more difficult. Additionally, tools available to visualize interactional behaviors within and surrounding the system of interest are lacking. Therefore, as identified in the OSD SBIR call, there is a substantial need for new tools to understand and visualize complex system behaviors during the design phase. This proposal outlines a method and software tool that can communicate the significance of component, subsystem, and system interactions to designers and generate workflows for the detection, evaluation, and remediation of unintended system interactions. Using the Marine Corps' Amphibious Assault Vehicle as a test case, our proposed technology will help users discover new information and enable them to understand how particular interactions are the result of system conditions and/or design actions. As delivered, our comprehensive displays of interactions will be of immediate use for integration within an Engineered Resilient Systems environment. Using our tool, problems are identified early on during the design process and can be eliminated through redesign, thereby allowing acquisition programs to mature and conclude faster and at a lower overall cost to the Government.

Intelligent Systems Technology, Inc.
12122 Victoria Ave
Los Angeles, CA 90066
Phone:
PI:
Topic#:
(310) 581-5440
Azad M. Madni
OSD12-ER1      Awarded: 7/12/2013
Title:VisualAnalytix™: Identification and Visualization of Interactions and their Consequences within Complex Systems
Abstract:In complex systems there tends to be greater likelihood of interactions and dependencies between components. These interactions and dependencies can give rise to unpredictable and often detrimental behaviors. To ameliorate this problem, there is a need for methods and tools that enable: detection, diagnosis, and visualization of component interactions; identification and alerting of undesirable interactions; and methods for circumventing undesirable interactions. Phase I of this effort is concerned with developing and demonstrating an interactive tool capable of helping users visualize interdependencies, test and analyze the interdependencies using diagnostic workflow routines, detect and visually depict undesirable interactions, and alert the user about detrimental interactions and their likely consequences.

Smart Information Flow Technologies, d/b/a SIFT
211 N 1st St. Suite 300
Minneapolis, MN 55401
Phone:
PI:
Topic#:
(435) 213-5776
Daniel Bryce
OSD12-ER1      Awarded: 5/2/2013
Title:STABILITY: Statistical Verification, Explanation Based Learning, and Interaction Testing for Complex Systems
Abstract:Complex systems are inherently difficult to analyze because they include multiple sub- systems that interact in stochastic and context-dependent ways. A particular challenge in designing such systems is to foresee the possible emergent interactions between the sub- systems. System repurposing makes the identification of interactions even more difficult, since the interactions vary along with the operational context. SIFT proposes to develop the STABILITY (Statistical Verification, Explanation Based Learning, and Interaction Testing for Complex Systems) system, a complex system design and analysis tool. STABILITY will support model based complex system development. STABILITY will combine human-guided workflow design, automated workflow analysis with statistical verification, succinct workflow analysis summarization with explanation based learning, and systematic workflow suite design with combinatorial interaction testing. STABILITY will provide a simple and effective framework in which to seek out, test, and understand interactions within complex systems.

Michigan Engineering Services, LLC
2890 Carpenter Road, Suite 1900
Ann Arbor, MI 48108
Phone:
PI:
Topic#:
(734) 429-7777
Geng Zhang
OSD12-ER2      Awarded: 5/22/2013
Title:Functional Allocation Trades Between Hardware and Software
Abstract:Design is the act of generating information that is used to make decisions or produce a product. Modern products are moving away from segregated disjointed systems and towards interdependent systems that utilize shared hardware and software resources (e.g. zonal power distribution, ethernet based controls, shared racks and mounts to facilitate advanced outfitting, etc.). Determining the appropriate balance between hardware and software configurations is an integral part of the design process. Modern design methods and tools still create and analyze designs as a set of disjointed systems, or in most cases, segregated analysis operations are organized by traditional technical domains. The concept of “shared resource” must become part of an integrated design process that considers allocation trade-off between hardware and software capabilities. Developing a software tool for balancing decisions when allocating investments in hardware and software during the development of complex DoD systems will have an impact on the lifecycle cost, the reliability, the robustness, and the utility of a new Defense system. Such a software tool will have the ability to select in an automated top-down approach the hardware/software configuration which will maximize the utility of the system through probabilistic analysis given a set of targeted performance characteristics. It will also be possible to use the new software in a manual bottom-up approach in order to assess the utility of a particular selection of hardware/software configuration made by the user. In either mode of operation, information will be presented in a graphical format for easy interpretation of the impacts that the selections have to the performance of the system. Such a research effort will be pursued by Michigan Engineering Services, LLC (MES) under the proposed SBIR program for developing a Decision Support toolkit for Hardware/Software allocation studies (DS toolkit for H/S).

VIStology, Inc
5 Mountainview Drive
Framingham, MA 01701
Phone:
PI:
Topic#:
(508) 788-5088
Jakub Moskal
OSD12-ER2      Awarded: 7/31/2013
Title:Functional Allocation Trades Between Hardware and Software
Abstract:It is expected that the US military forces will be facing many new, unexpected types of mission – from combat operations to humanitarian assistance and disaster recovery (HA/DR). To be able to accomplish such missions, the military will require “resilient systems” of all kinds – digital, electrical and mechanical. The development of such new systems requires new design tools that can support the designer in exploring solution spaces and analyzing the suitability of a specific solution for a specific mission or task, as well as for a class of potential future missions and tasks. Such a capability is referred to as “tradespace exploration”. This project will develop a prototype of a tool to support the designer of a resilient system in allocating functionalities to either hardware or software. The tool will be based on the principles of optimization and will provide a rationalization for the particular selections of the allocations. The tool will utilize models expressed in both UML and in semantic languages, like OWL. The designer will be able to enter problem formulations, descriptions of functions, systems and components. Additionally, this project will analyze and formalize various allocation problems and investigate algorithms for specific problems. VIStology will be collaborating with Raytheon, especially in the development of use cases and scenarios, as well as in the evaluation of the tool.

WW Technology Group
4519 Mustering Drum
Ellicott City, MD 21042
Phone:
PI:
Topic#:
(410) 418-4353
Chris J. Walter
OSD12-ER2      Awarded: 8/23/2013
Title:Functional Allocation Trades Between Hardware and Software
Abstract:Design, analysis, and assessment of multiple alternatives is important to ensuring the development and fielding of DoD systems that face increasingly dynamic and rapid changes in missions. New methods and tools are needed that compensate and recover from disruptions, adapt to dynamic environments, and rapidly deliver new solutions. A key goal is ensuring that engineering programs maximize utility and represent best value for the investment across a breadth of potential missions in joint operational environments. In this Phase I SBIR, we will develop a software tool with an approach for allocating system functions to implementations of hardware or software. The design studies will be evaluated with quantitative assessment techniques to determine the strengths and weaknesses of each allocation options. This will be provided in a context that supports multiple abstraction layers and decomposition views from a hierarchically increasing view (e.g. component, subsystem, system, and system-of-systems). The software tool uses a design study method that enables comparative assessments to be made between allocations of the same function to hardware and software implementations. The results will be summarized in a dashboard display to the decision maker with graphical visualizations supported by quantitative assessments of factors evaluations.

Sonalysts
215 Parkway North P.O. Box 280
Waterford, CT 06385
Phone:
PI:
Topic#:
(540) 663-9034
Stephen Dorton
OSD12-HS1      Awarded: 5/14/2013
Title:Human Computer Interfaces for supervisory control of Multi-mission, Multi-Agent Autonomy
Abstract:We propose to develop a collaborative control user interface that will enable operators to work more effectively with highly autonomous vehicles by engaging in a dialog to exchange ideas, ask questions, and resolve differences. We will develop a concept of operations for the collaborative control user interface, and we will create specific, representative scenarios describing situations and events that benefit from collaborative control. We will develop a preliminary ontology that describes the semantics and pragmatics of the information exchange between the operator and the vehicle. We will develop a software architecture that specifies the capabilities, algorithms, knowledge requirements, and interfaces of each software module. We will develop a limited software prototype that illustrates key elements of our approach to demonstrate its utility and feasibility. Finally, we will develop autonomous vehicle evaluation metrics, paying special attention to the efficiency and effectiveness of the operator-vehicle collaboration.

Stottler Henke Associates, Inc.
1670 South Amphlett Blvd. Suite 310
San Mateo, CA 94402
Phone:
PI:
Topic#:
(650) 931-2700
Emilio Remolina
OSD12-HS1      Awarded: 5/14/2013
Title:Intelligent Collaborative Control User Interfaces to Autonomous Vehicles
Abstract:We propose to develop a collaborative control user interface that will enable operators to work more effectively with highly autonomous vehicles by engaging in a dialog to exchange ideas, ask questions, and resolve differences. We will develop a concept of operations for the collaborative control user interface, and we will create specific, representative scenarios describing situations and events that benefit from collaborative control. We will develop a preliminary ontology that describes the semantics and pragmatics of the information exchange between the operator and the vehicle. We will develop a software architecture that specifies the capabilities, algorithms, knowledge requirements, and interfaces of each software module. We will develop a limited software prototype that illustrates key elements of our approach to demonstrate its utility and feasibility. Finally, we will develop autonomous vehicle evaluation metrics, paying special attention to the efficiency and effectiveness of the operator-vehicle collaboration.

Traclabs, Inc.
100 Northeast Loop 410 Suite 520
San Antonio, TX 78216
Phone:
PI:
Topic#:
(281) 461-7886
David Kortenkamp
OSD12-HS1      Awarded: 5/15/2013
Title:Autonomy Management Platform
Abstract:Autonomous vehicles are becoming more and more integral to our nation's defense. A DoD object is to free humans from direct control and move towards supervisory control of multiple, autonomous assets. Achieving this objective will require significant advances in decision support and supervisory control concepts. Current operator interfaces require continuous attention from the operator to a single unmanned vehicle. In fact, many require the attention of several operators. In addition, designing and building the decision support components often requires significant expertise in computer science, artificial intelligence, and software engineering. This makes the cost of developing, maintaining, and modifying these systems exorbitant. Our approach is to design reusable, ontology-driven decision support tools that can be configured into an Autonomy Management Platform (AMP) that can assist an operator in controlling and directing several unmanned vehicles at the same time. We also propose an ontology editor that lets non-computer scientists create and update the assets, capabilities, tasks, and environment of the various unmanned vehicles. These ontologies are used by the decision support tools, which allows for flexibility and extensibility of AMP. The benefits will be reduced cost of operations and increased productivity of the operators while saving time and money in the design, deployment, and maintenance of decision support systems for multi-asset coordination.

Aptima, Inc.
12 Gill Street Suite 1400
Woburn, MA 01801
Phone:
PI:
Topic#:
(781) 935-3966
Stacy Pfautz
OSD12-HS2      Awarded: 2/26/2013
Title:SONIC – Sensor Operations via Naturalistic Interactive Control
Abstract:Intelligence Surveillance and Reconnaissance (ISR) requires the ability to navigate and interpret mounds of data to produce actionable decisions. The ever-changing need to incorporate new capabilities into daily operations has increased the complexity of this task in recent years. Through the ICE Box environment, AFRL is redefining a ConOps for future ISR operations and redesigning how analysts interact with ISR technologies. However, the ICE Box human-machine interface does not yet provide a “naturalistic” experience to its users. To support this need, the Aptima team proposes to develop SONIC (Sensor Operations via Naturalistic Interactive Control), a multi-modal user interaction framework optimized for use within highly immersive workspaces such as ICE Box, to provide a naturalistic way for analysts to control and receive information from remote networks of sensors. SONIC will emphasize three primary design criteria: (1) incorporating relevant human factors constructs into the interaction design framework; (2) intelligent fusion of multi-modal user inputs; and (3) integrating contextual information into the user interface. Ultimately, the objective of SONIC is to enable analysts to collaborate and provide mission support from remote locations more effectively without an increase in workload or a decrease in performance.

Design Interactive, Inc.
1221 E. Broadway, Suite 110
Oviedo, FL 32765
Phone:
PI:
Topic#:
(407) 706-0977
Kelly Hale
OSD12-HS2      Awarded: 2/19/2013
Title:Ecological Gesturing Ontology (EGO)
Abstract:Intelligence from strategic areas of interest comes from numerous sources, such as Unmanned Aerial Vehicles (UAVs) and Rapid Aerostat Initial Deployment (RAID) cameras, and in various formats, such as imagery, video, audio, and text. Searching through the information for analysis can be challenging, particularly when the intelligence analyst is not co-located within the conflict area. Design Interactive, Inc., along with the Electrical Engineering and Computer Science (EECS) group at the University of Central Florida, proposes the Ecological Gesturing Ontology (EGO), a primarily naturalistic gesture-based control interface for fully-immersive, synthetically-augmented displays. EGO allows the teleworking intelligence analyst to stand within a fully-immersive synthetically-augmented environment and, through ecologically designed naturalistic gesturing and vocal commands, purposefully investigate integrated intelligence data within the environment by exploring, searching for, finding, and reviewing sensor data. Further, the naturalistic human-machine interaction will incorporate a graphical interface that enables efficient interaction with and transitions between various modes of information presentation – GEOINT, HUMINT, SIGINT, etc. EGO shall be the first naturalistic human machine interface design to support the best practices of multimodal information presentation across multiple interaction modalities that will increase presence, decrease workload, and support decision-making.

Hadron Industries, Inc.
90 Airport Road
Concord, NH 03301
Phone:
PI:
Topic#:
(855) 267-4253
Frank Tanner
OSD12-HS2      Awarded: 2/28/2013
Title:Naturalistic Operator Interface for Immersive Environments
Abstract:"Remotely sensed data has an inherent physical detachment from the environment being sensed. At a high-level, we propose leveraging Oblong Industries’ revolutionary g-speak platform to restore fully embodied situational awareness through multimodal interaction within the ICEbox environment.The g-speak spatial operating environment (SOE) is the world’s first fully-embodied, spatially aware computing environment, and the result of nearly 25 years of research and development into novel methods of human-computer interaction. The g-speak system’s conceptual debut was in the movie Minority Report and recently appeared in the Iron Man films. What began as a movie concept has developed into a spatially aware, networked OS, designed from the ground-up as a whole-body, human- computer interaction platform. The g-speak SOE is designed to help programmers write applications that use gestural input, function well on large screens and for simultaneous users, work across multiple computers and screens, and can be built from loosely-coupled small programs. By design, g-speak understands the relationships between the operator and the environment. By using a variety of hand gestures, as well as interaction with physical objects and/or mobile devices, g-speak is uniquely capable of restoring embodiment to the task of computing in general, and immersive situational awareness in particular."

Unova Technologies
9015 N. Cobre Drive
Phoenix, AZ 85028
Phone:
PI:
Topic#:
(480) 363-5109
Jeffrey A. Getzlaff
OSD12-HS2      Awarded: 2/22/2013
Title:Naturalistic Operator Interface for Immersive Environments
Abstract:This proposal addresses the significant need for supervisory control of sensor networks within a fully-immersive synthetic environment with novel Human-Machine Interfaces (HMIs). A methodology and process to design a Synthetic Environment Machine Interface System (SEMIS) with multi-modal inference processing based on gestures and speech is detailed. The Phase 1 Work Plan employs the Rational Unified Process and AGILE software methodologies to ensure a focus on a supervisory operator’s needs and system goals. Examples of innovative proposed functionality include: a collaborative unmanned sensor group, sensor action ring overlay, HMI sensor toolkits, gesture table, 3D holograms, temporal timeline, health and status displays and data product tools. A cognitive analysis plan is utilized for analysis of problem domain functionality with respect to operator workload, varying levels of automation, and portrayal of information for sensor monitoring and re- tasking. A software testbed is also created for analysis and testing of proposed algorithms and HMI visualizations. At the end of Phase I, results of the research of are presented coupled with a limited demonstration of the SEMIS with the AFRL’s ICEbox in a supervisory operator mission scenario. Market segments are defined for commercialization, including PED systems, UxV ground control stations, medical imaging, and Cyber Warfare markets.

Charles River Analytics Inc.
625 Mount Auburn Street
Cambridge, MA 02138
Phone:
PI:
Topic#:
(617) 491-3474
Camille Monnier
OSD12-HS3      Awarded: 5/22/2013
Title:Taxiing Operations via Gesture Understanding (TOPGUN)
Abstract:Unmanned Aerial Systems (UAS) play an increasingly important role in many military scenarios. For the Navy to incorporate UAS—such as those currently being developed under the Unmanned Combat Air System Carrier Demonstration (UCAS-D) program—into its critical missions without impacting the current sortie rate, an efficient protocol is needed for handling UAS on aircraft carrier decks. We propose to design and develop a system for Taxiing OPerations via Gesture UNderstanding (TOPGUN). The fundamental premise of our approach is to use the same communication modality (i.e., gestures) for controlling UAS as for piloted aircraft. Our solution uses on-board cameras to track flight deck directors and to recognize their gestures without requiring modifications to operator equipment or to the carrier deck. The underlying software will build on state-of-the-art video-based human tracking, pose estimation, and gesture recognition technologies previously developed by our team to produce an efficient and reliable system for gesture-based control of UAS during carrier deck operations.

Physical Optics Corporation
Electro-Optics Systems Division 1845 West 205th Street
Torrance, CA 90501
Phone:
PI:
Topic#:
(310) 320-3088
Ofir Garcia
OSD12-HS3      Awarded: 5/23/2013
Title:Insect-Inspired Non-Imaging Gesture Human Tracking Array Optics
Abstract:To address the OSD need for a minimally intrusive technology that supports aircraft carrier operator gesture recognition, Physical Optics Corporation (POC) proposes to develop a new Non-Imaging Gesture Human Tracking Array Optics (NIGHT-AO). This proposed device is based on a novel, low size weight and power (SWAP), insect-inspired dual-band (visible-infrared) optical array design that utilizes POC-developed mature components and COTS components. The innovation in optical array configuration, angular resolution, embedded processing network algorithm and speed (at low SWAP) will enable the device to rapidly recognize, acknowledge, and learn new commands used in standard carrier operations. As a result, this device offers seamless integration of upcoming unmanned air systems (UAS) into current carrier operations, and maintains or improves current sortie rates, directly addressing upcoming UAS acquisition program requirements. In Phase I, POC will provide a feasibility study for a natural dialogue-based gesture-recognition human control interface for UAS in carrier-based operations. A final report will be delivered including system performance metrics and plans for Phase II. Additionally, POC will demonstrate the feasibility of NIGHT-AO by building a proof-of-concept prototype. In Phase II, POC plans to build an improved prototype system to be tested in a laboratory simulating the aircraft-carrier CONOPS environment.

Systems Technology, Inc.
13766 S. Hawthorne Blvd.
Hawthorne, CA 90250
Phone:
PI:
Topic#:
(310) 679-2281
Edward Bachelder
OSD12-HS3      Awarded: 5/23/2013
Title:Natural Dialogue – based Gesture Recognition for Unmanned Aerial System Carrier Deck Operations
Abstract:"Remotely sensed data has an inherent physical detachment from the environment being sensed. At a high-level, we propose leveraging Oblong Industries’ revolutionary g-speak platform to restore fully embodied situational awareness through multimodal interaction within the ICEbox environment.The g-speak spatial operating environment (SOE) is the world’s first fully-embodied, spatially aware computing environment, and the result of nearly 25 years of research and development into novel methods of human-computer interaction. The g-speak system’s conceptual debut was in the movie Minority Report and recently appeared in the Iron Man films. What began as a movie concept has developed into a spatially aware, networked OS, designed from the ground-up as a whole-body, human- computer interaction platform. The g-speak SOE is designed to help programmers write applications that use gestural input, function well on large screens and for simultaneous users, work across multiple computers and screens, and can be built from loosely-coupled small programs. By design, g-speak understands the relationships between the operator and the environment. By using a variety of hand gestures, as well as interaction with physical objects and/or mobile devices, g-speak is uniquely capable of restoring embodiment to the task of computing in general, and immersive situational awareness in particular."

Charles River Analytics Inc.
625 Mount Auburn Street
Cambridge, MA 02138
Phone:
PI:
Topic#:
(617) 491-3474
Sean Guarino
OSD12-IA1      Awarded: 2/25/2013
Title:Toolkit for Managing Evaluation and Testing for Red Team Investigations of Cyber Security (METRICS)
Abstract:Adversaries have become increasingly proficient at cyber attacks against our military’s command and control (C2) infrastructure. Maintaining security requires high-fidelity assessments of software services, often implemented as cyber Red Team exercises in which Systems Under Test (SUTs) are subjected to attacks designed to evaluate defensive capabilities. These tests produce massive amounts of data with subtle patterns and effects that can be difficult to interpret post-experimentally, let alone in real time where these effects would enable more thorough, dynamic, and realistic testing of SUT security. To address this need, we propose to design and demonstrate a toolkit for Managing Evaluation and Testing for Red Team Investigations of Cyber Security (METRICS). The METRICS toolkit includes four key components: (1) a library of contextualized metrics that incorporate a full understanding of system and attack implications to support real-time assessment of the SUT cyber defense; (2) intuitive authoring tools for customizing and developing metrics for evolving SUTs, attacks, and experiment needs; (3) adaptable and adaptive visualizations that present analysis results in manner that ensures observability of critical patterns during ongoing experiments; and (4) a collection harness that employs COTS packet sniffers and agent-based data collection tools to non-intrusively collect data for analysis.

Global InfoTek, Inc
1920 Association Drive Suite 200
Reston, VA 20191
Phone:
PI:
Topic#:
(703) 652-1600
Ray Emami
OSD12-IA1      Awarded: 2/25/2013
Title:Coral Viz
Abstract:Global InfoTek, Inc. will research methods for developmental testing of cyber attacks on enterprise applications and services. The research will focus on visualization, analytics and data capture approaches for determining the impact of cyber attacks on Service Oriented Architecture (SOA) based services and applications. The resulting research will create prototype tools that can assist development teams in identifying security vulnerabilities, performance inhibitors and robustness of services that would be deployed in conjunction with a SOA framework. The initial challenges will address how to create a flexible visualization platform that can show the impact of a cyber attack on multiple layers of the Opens System Interconnection (OSI) stack from hardware performance to network topologies to security services to data services. This visualization will be accompanied by an analytical engine that can make meaning of the data captures from enterprise sensors and provide it to developers in a manner that is meaningful and useful. These tools will be specifically designed to adapt to the rapid addition and subtraction of SOA services and to provide repeatable tests that developers can use to test, debug and validate service and application performance.

Intelligent Automation, Inc.
15400 Calhoun Drive Suite 400
Rockville, MD 20855
Phone:
PI:
Topic#:
(301) 294-4633
Song Luo
OSD12-IA1      Awarded: 2/25/2013
Title:Hermes: A Visualized and Automated Cyber Security Assessment Toolkit
Abstract:IAI and its transition partners, Raytheon BBN Technologies and Raytheon Integrated Defense Systems, propose the Hermes framework and toolkit for cyber security assessment and evaluation tests, which facilitates test exercises in managing, processing, and analyzing large logging datasets from distributed and service-centric computing test bed. Another major capability of this toolkit is that it gains an aggregate system view before, during, and post assessment exercises, and delivers analytical results on various performance metrics qualitatively, quantitatively, and visually. Hermes is also capable of predictive analysis based on network topology, attack progression, and host/application status. It provides a holistic DVR-like visualization which captures not only the hardware and software resources in the test bed but also attacking and defending progresses, and allows testers to record and repeatedly replay assessment exercises. The Hermes framework will greatly help cyber testers obtain the insight of attack/defense mechanism and Systems Under Test (SUT).

21CT, Inc.
6011 West Courtyard Drive Bldg 5, Suite 300
Austin, TX 78730
Phone:
PI:
Topic#:
(512) 682-4730
Jonathan Mugan
OSD12-IA2      Awarded: 2/22/2013
Title:SATPAM - Self Awareness Through Predictive Abstraction Modeling
Abstract:Current computer systems are dumb automatons; their blind execution of instructions makes them open to attack. Their inability to reason means they don’t consider the larger, constantly changing context outside their immediate inputs. Their nearsightedness is particularly dangerous because, in our complex systems, it is difficult to prevent all exploitable situations. Additionally, the lack of autonomous oversight of our systems means they are unable to fight through attacks. Keeping our adversaries completely out of our systems may be an unreasonable expectation, and our systems need to adapt to attacks and other disruptions to achieve our objectives. What is needed is an autonomous controller within the computer system that can sense the state of the system and reason about that state. 21CT proposes SATPAM, which uses prediction to learn abstractions that allow it to recognize the right events at the right level of detail. These abstractions allow SATPAM to break the world into small, relatively independent, pieces that allow employment of existing reasoning methods.

ATC - NY
33 Thornwood Drive, Suite 500
Ithaca, NY 14850
Phone:
PI:
Topic#:
(607) 257-1975
Hajime Inoue
OSD12-IA2      Awarded: 3/7/2013
Title:VMCIS: A Cognitive Immune System for Virtual Machine-based Mission Critical Applications
Abstract:Cyber systems remain susceptible to attack and compromise despite the best security efforts. There are two growing trends in the research community to address this: a paradigm of prioritizing missions over individual network resources, and a push for resilient systems capable of fighting through cyber attacks. ATC-NY will develop VMCIS, the Virtual Machine Cognitive Immune System, which combines both approaches. VMCIS constantly monitors systems at many execution levels and automatically takes corrective actions when necessary. VMCIS reasons about system state using a probabilistic, hierarchical logic engine that is built around a model of mission-success. It directs corrective actions using a System Controller that can manipulate both the hypervisor to checkpoint, rollback, and migrate VMs, and the operating system to install, remove, restart, or kill applications. VMCIS also provides APIs so developers can add customized instrumentation and remediation actions.

GrammaTech, Inc
531 Esty Street
Ithaca, NY 14850
Phone:
PI:
Topic#:
(949) 573-8814
Brad Arant
OSD12-IA2      Awarded: 3/13/2013
Title:Multi-Abstractions System Reasoning Infrastructure toward Achieving Adaptive Computing Systems
Abstract:"The complexity of modern computer systems has grown to the point of stressing human ability to understand their behavior completely. The sheer number of software components (and the myriad interactions between them) that are present on a single desktop computer presents a difficult security challenge that continues to confound modern protection technologies. Every day, new exploits are created that take advantage of obscure combinations of software bugs and unexpected behavior to sidestep existing defenses. Response is slow, requiring human effort to diagnose and develop new counter-measures to each new threat.GrammaTech envisions a new paradigm for building system security that endows the computer, itself, with the tools to diagnose new attacks, reason about their impact to the system, and implement countermeasures in an automatic fashion. Autonomy- oriented computation offers the potential to allow complex computer systems to police themselves, detecting intrusion, performing self-healing, and directly countering cyber threats."

Intelligent Automation, Inc.
15400 Calhoun Drive Suite 400
Rockville, MD 20855
Phone:
PI:
Topic#:
(301) 294-5218
Peng Xie
OSD12-IA2      Awarded: 2/27/2013
Title:SAM: A Self-adaptive Monitoring System Architecture
Abstract:Self-adaptive monitoring systems are highly desirable and demanded in military and civilian application scenarios. In this effort, we leverage our in-house machine learning framework, ABMiner and ontological knowledge representation workbench to address the challenging problem, and propose the self-adaptive monitoring architecture, called SAM. In SAM, we mature data mining techniques to set up a multi-feature model for the monitored application. The multi-feature monitoring model can be used to monitor the execution of the application online. ABMiner tool allows us to use various machine learning techniques, including ensembles, to select the most significant semantic features to characterize the monitored application. Moreover, we will enhance ABMiner by integrating the weighted ensemble methods to automatically adjust to drifting features inherited in software applications. Additionally, we leverage our work on ontological knowledge representation techniques to represent the execution of an application program in an understandable way to enable human experts to adjust the automatic reasoning system. Finally, we will integrate the proposed techniques in a workable self-adaptive monitoring prototype capable of checking the execution of dynamically evolving applications.

Charles River Analytics Inc.
625 Mount Auburn Street
Cambridge, MA 02138
Phone:
PI:
Topic#:
(617) 491-3474
Erik Thomsen
OSD12-IA3      Awarded: 6/10/2013
Title:Framework for Computing Assurance Measurements based on Evolving Resilience Metrics (FRAMER)
Abstract:As command and control, doctrine, and tactics evolve with the cyber network at its core, mission-critical tasks increasingly rely on the integrity and responsiveness of the network and other cyber assets. While this trend has acted as a tremendous force multiplier and OPTEMPO accelerator, two major concerns arise: (1) network usage demands will soon outstrip the capabilities of a resource-constrained network (e.g., download requests for UAV data exceeding available bandwidth); and (2) the growing capabilities of adversaries to wage cyber warfare will disrupt networks. To assess mission assurance based on the resilience and criticality of cyber assets, we propose to design and demonstrate a Framework for Computing Assurance Measurements based on Evolving Resilience Metrics (FRAMER) that measures mission assurance through cyber asset resilience and criticality metrics. FRAMER introduces a new set of criticality and resilience metrics by maintaining a constantly evolving set of mappings between cyber assets and prioritized mission elements. Together, the mappings and relationships drive the derivation of criticality and resilience measures for each cyber asset. FRAMER assesses mission assurance for each mission by aggregating the criticality and resilience of each cyber asset mapped to a collection of missions.

Intelligent Automation, Inc.
15400 Calhoun Drive Suite 400
Rockville, MD 20855
Phone:
PI:
Topic#:
(301) 294-5215
Yi Cheng
OSD12-IA3      Awarded: 5/7/2013
Title:CREACT: Advanced Network Security Metrics for Cyber REsilience and Asset CriTicality Measurement in Mission Success
Abstract:Cyber assets usually support missions with different priorities. How to effectively evaluate the health of a network and its ability to achieve the overall mission is critical to ensure mission success. In this effort, Intelligent Automation, Inc. proposes “CREACT”, a set of advanced network security metrics for cyber resilience and asset criticality measurement in a large- scale and dynamic network environment. Essentially, advanced valued-based goal models and efficient mission-to-asset mapping, resource allocation, vulnerability assessment, threat analysis and impact mitigation techniques will be developed for cyber resilience and asset criticality modeling, evaluation and measurement. The developed technologies will be implemented into an integrated software toolkit for comprehensive cyber security analysis and network defense to achieve mission success.

Intelligent Automation, Inc.
15400 Calhoun Drive Suite 400
Rockville, MD 20855
Phone:
PI:
Topic#:
(301) 294-4251
Justin Yackoski
OSD12-IA4      Awarded: 6/4/2013
Title:Recognition of New Advanced Threats Using Purpose and Network Correlators (RAPCOR)
Abstract:Cyber intrusion and anomaly detection techniques suffer from their reliance upon the presence of known malicious signatures or unusual conditions that warrant further investigation. The use of signature-based detection cannot effectively eliminate false negatives when dealing with advanced persistent threats (APTs). Furthermore, current detection tools incur very high false positives. To address such challenges, Intelligent Automation, Inc. (IAI), along with Prof. Guofei Gu from Texas A&M University, proposes to develop novel non-signature based APT detection algorithms that allow proper identification, prioritization, and understanding of attacks. The key innovation is to place detectors within the network and individual hosts to provide real-time purpose and correlation inputs and then use this information combined with network-specific knowledge to create a dynamic “spider web”-like set of threads that, when touched by a given alert, allow the immediate identification of the context surrounding the alert and thus the automatic calculation of the alert’s legitimacy and severity. The result is that much of the follow-up investigation of each alert is shifted into the attack prioritization process, allowing this context to correctly prioritize alerts. The burden on operators is thus reduced both by significantly improved prioritization and by providing a contextual picture of each potential attack identified.

Numerica Corporation
4850 Hahns Peak Drive Suite 200
Loveland, CO 80538
Phone:
PI:
Topic#:
(970) 612-2333
Randy Paffenroth
OSD12-IA4      Awarded: 6/11/2013
Title:Novel Detection Mechanisms for Advanced Persistent Threats
Abstract:Department of Defense (DoD) operations are supported by a global network of computers, sensors, and equipment that is continually at risk of being breached by adversaries. Despite heavy investments in security and cyber defense, the ubiquity and interconnectedness of DoD equipment leave open the possibility of intrusion through a myriad of means including advanced persistent threats (APTs). Such threats take many forms, such as Trojans, worms, spear-phishing, and viruses, all of which could prove detrimental to the war-fighter if not discovered. Unfortunately, the ``base rate fallacy' places fundamental limits on the performance of detection algorithms in the cyber-defense context. Are there any directions left in which to tackle this important problem? We would claim the answer to that question is a resounding ``yes', and modern techniques in sensor fusion, multiple hypothesis testing, and compressed sensing lead to algorithms with quite advantageous properties. These methods have all paid large dividends in other problem domains, such as medical studies and mathematical finance, but have not yet seen their full bloom in cyber-defense problems, a deficit we hope to remedy herein. In particular, a judicious choice of sensors and sensor fusion methodologies provide promising paths for improving the state of the art.

Paradigm Shift International
2051 Lama Mountain Box 289
Questa, NM 87556
Phone:
PI:
Topic#:
(575) 586-1536
Rick Dove
OSD12-IA4      Awarded: 9/6/2013
Title:Novel Detection Mechanisms for Advanced Persistent Threat
Abstract:This project employs a massively parallel, low cost, low power, associative-memory pattern detection processor soon-to-market by a major semiconductor producer. Phase 1 will use a microprocessor emulator to develop, test, and analyze “very large scale anomaly detectors” (developed under a prior SBIR project) organized in a 3-level hierarchical sense-making architecture of spatial, temporal, and correlative pattern detectors – for employment at network endpoints. A fourth level in the sense-making hierarchy will be deferred until Phase 2, and provide cross-endpoint network-wide correlative pattern detection. The Phase 1 project has three principle objectives: 1) to establish performance and values of the very large scale anomaly detectors for detecting zero-day and advanced persistent threat attacks, and 2) to develop a semi-supervised learning process that converges on a sparse but sufficiently optimal pattern dictionary for each of the three levels in the hierarchy. and 3) to demonstrate capability to discover previously unseen attacks with high true positives and low false positives.

Charles River Analytics Inc.
625 Mount Auburn Street
Cambridge, MA 02138
Phone:
PI:
Topic#:
(617) 491-3474
Joseph Gorman
OSD12-IA5      Awarded: 5/7/2013
Title:Feed Aggregation for Threat Evaluation Service (FATES)
Abstract:CNDSP analysts protect US computer networks, detect threat occurrences on those networks, and respond to threat incidents. The adage “know thy enemy” is as true for network defense as it is for defense of any asset. A robust computer network defense requires current knowledge of possible and actual threats and knowledge of conditions within the network. However, analysts must invest valuable time manually gathering and processing threat descriptions, which reduces the time available for understanding the nature of new threats, preparing responses to those threats, monitoring network conditions for threat occurrences, and responding to threat incidents. Approaches are needed that automate and make routine collection of threat information, aggregating collected information into comprehensive threat descriptions, assessing the risk severity posed to US networks, and generating meaningful I&W for a future or occurring threat. We propose a Feed Aggregation and Threat Evaluation Service (FATES) that will: (1) automatically collect network threat information from multiple threat information feeds, (2) aggregate information from these feeds to produce a timely comprehensive threat description, (3) provide web services that distribute threat descriptions to CNDSP analysts displays, and (4) provide a web service that distributes advanced I&W for CNDSP analysts.

InferLink Corporation
2361 Rosecrans Avenue, Suite 348
El Segundo, CA 90245
Phone:
PI:
Topic#:
(310) 944-4813
Greg Barish
OSD12-IA5      Awarded: 6/25/2013
Title:ThreatRank: aggregating and prioritizing network threats
Abstract:The need to defend computer networks becomes increasingly vital as the volume of data generated by both man and machine continues to explode. However, the capability of rapidly analyzing large numbers of heterogeneous threat sensors (feeds) for various security outposts has become its own challenge. Ideally, one would like to aggregate these threat feeds and prioritize incidents at a global view, based on the entire context that is observable. However, this is not trivial; feeds vary in terms of how they report similar information, and there is no central aggregation point to cluster similar threats. For this SBIR, we propose ThreatRank, a system for rapidly and automatically integrating, aggregating, and ranking data from these feeds, to organize and simplify the global view of current threats, so that situational awareness can be maximized. Our approach builds on our significant experience with AI and statistically-based approaches to intelligent data integration and prediction, offering a way to handle the problem in a scalable, maintainable manner.

Intelligent Automation, Inc.
15400 Calhoun Drive Suite 400
Rockville, MD 20855
Phone:
PI:
Topic#:
(301) 294-4634
Jyotirmaya Nanda
OSD12-IA5      Awarded: 5/22/2013
Title:An Integrated Threat feed Aggregation, Analysis, and Visualization (TAAV) Tool for Cyber Situational Awareness
Abstract:Cyber security analysts are inundated with heterogeneous threat feeds created by different kinds of cyber security monitoring tools such as Snort, Nessus, Symantec etc. There is a need for streamlining threat analysis to help operators focus on prompt identification and comprehension of security threats early on. To address this critical need, Intelligent Automation Inc. (IAI), with Prof. Peng Liu from Penn State University, is proposing an integrated tool for heterogeneous and multi-structured Threat feed Aggregation, Analysis, and Visualization (TAAV). The proposed framework will execute automated expert and data driven multi-dimensional and time correlated threat feed analysis that will categorize threat feeds into “threat baskets”. The identified relevant threat baskets will be prioritized according to their perceived scale of vulnerability by performing latent threat association detection and alert correlation over ongoing as well as historical attack information. The result of the automated analysis will be presented in an easily understandable display with accompanying maps and charts. An empirical study will be performed on the Phase I prototype using a scalable test bed. TAAV will be developed end-to-end in Java using open source tools (e.g. Quartz scheduler, Spring Batch, Talend, JBoss Drools, Accumulo).

Intelligent Automation, Inc.
15400 Calhoun Drive Suite 400
Rockville, MD 20855
Phone:
PI:
Topic#:
(301) 294-4454
Sohraab Soltani
OSD12-IA6      Awarded: 5/22/2013
Title:RADAR: A Comprehensive and Dynamic Framework toward Real time Network Traffic Resiliency
Abstract:This effort proposes a comprehensive framework to design and develop an intelligent, dynamic and real time traffic monitoring/filtering system to identify and covertly divert suspicious/malicious traffic (within an enterprise network) to specific locations for further analysis and to ensure resilient network traffic boundary. The proposed approach exploits existing algorithms/technologies (e.g., commercial traffic monitoring/filtering tools, threat detection algorithms, commercial routers configurations) to develop a complete traffic resilience enabling system that dynamically monitors, detects and intelligently redirects suspicious traffic to isolated locations in the network in a real time manner. We refer to the proposed framework as Realtime, Adaptive and Dynamic trAffic Resilient, the RADAR framework. Using IAIfs experimental network testbed, we will study the feasibility of the proposed RADAR framework over a network comprising of hybrid sets of routers. We will integrate software routers, routers which support open source firmware, such as OpenWrt or DD-WRT, and other widely used commercial routers such as CISCO, Juniper and Linksys into our implementation. A preliminary prototype will be provided to demonstrate our Dynamic Traffic Resilience mechanisms.

Physical Optics Corporation
Applied Technologies Division 1845 W. 205th Street
Torrance, CA 90501
Phone:
PI:
Topic#:
(310) 320-3088
Alexander Milovanov
OSD12-IA6      Awarded: 7/19/2013
Title:Threats Redirection and Analysis System
Abstract:To address the OSD need for an innovative model and a dynamic system of cyber threat identification coupled with a means to redirect suspicious or malicious network activity to independent locations for investigation, Physical Optics Corporation (POC) proposes to develop a new Threats Redirection and Analysis System (TRAS). TRAS integrates POC- developed network protection model with existing BGP FLOWSPEC. TRAS monitors network traffic and dynamically detects attacks despite signature. It correlates and aggregates different threat data and uses FLOWSPEC to covertly redirect detected malicious traffic to disparate location for DoD to perform organized analysis of advanced persistent threat. TRAS’s rule-based intrusion detection mechanism enables detecting stealth network reconnaissance, low frequency attacks, and currently undetectable new generation of cyber attacks such as highly distributed and coordinated attacks. In Phase I, POC will define a plan to develop TRAS’s model and ruleset and demonstrate the feasibility of concept by assembling and testing a prototype. In Phase II, POC plans to develop an advanced prototype to demonstrate the capability to directly manipulate the path of network activity and hop count using IPv6 and reliably protect networks against existing and future attacks.

Black River Systems Company, Inc.
162 Genesee Street
Utica, NY 13502
Phone:
PI:
Topic#:
(315) 732-7385
Dale Klamer
OSD12-LD1      Awarded: 2/25/2013
Title:Autonomous Sensing and Deciding Framework Processor
Abstract:Our objective is to develop an innovative cognitive knowledge-aided information processing framework to take very high rate intelligence data streams over wide areas and autonomously highlight AOIs and targets for the image analyst without a priori knowledge of the area or location of the individual high interest targets. We will design, implement, test, and demonstrate an initial multi-agent autonomous sensing and deciding framework, including the development of knowledge bases, detecting anomalies between a real-time scene and the knowledge bases, mining the knowledge bases for new patterns and relationships, monitoring Areas of Interest specified by the analyst, and provide automated machine learning and ranking of anomalies within the multi-agent framework. Based on traffic pattern analysis and pattern of life, we derived knowledge bases and apply sound statistical analysis to detect changes between the current real-time scene and the derived knowledge bases. Through the use of Relevance Vector Machines, machine learning agents learn the important characteristics of AOIs and individual targets.

Stottler Henke Associates, Inc.
1670 South Amphlett Blvd. Suite 310
San Mateo, CA 94402
Phone:
PI:
Topic#:
(650) 931-2700
Richard Stottler
OSD12-LD1      Awarded: 2/25/2013
Title:A framework for fielding Artificial Intelligence techniques on High Performance Computing (HPC) Hardware
Abstract:The ultimate goal of this proposed effort is to improve the ability to determine suspicious radar track behavior and suspicious areas to focus attention on. We propose a framework for representing the knowledge and human-quality reasoning required to process large quantities of radar data, transforming it into a form that can be efficiently executed in real time on an HPC system, and then efficiently executing it given the actual dynamic tactical situation. In addition to being scalable up to a large number of pixels and objects, it should also be adaptable both in the long term to different HPC configurations and in real time to different dynamic computational loads. In Phase I we will study the urban environment and resulting radar data; determine the tactical reasoning and available information that can be applied to knowledge-based frame- to-frame radar track correlation and vehicle behavior analysis; elaborate the heuristics and algorithms for learning traffic patterns, normalcy, and other knowledge from past data; develop techniques for the automatic translation to a form for HPC hardware; develop real- time HPC resource scheduling techniques; prove the feasibility through prototype development, experimental testing with real radar data, and demonstration; and develop the Phase II system design.

Toyon Research Corp.
6800 Cortona Drive
Goleta, CA 93117
Phone:
PI:
Topic#:
(805) 968-6787
Charlene S. Ahn
OSD12-LD1      Awarded: 2/25/2013
Title:Autonomous Sensing and Deciding Framework Processor
Abstract:Toyon Research Corporation proposes to develop innovative algorithms for screening large volumes of intelligence data to autonomously highlight areas of interest for the research analyst. Toyon’s dual-level architecture will at the first level employ advanced data fusion and pattern recognition technologies to obtain long-term persistent tracks from SAR information data products. The fusion algorithms will provide persistent feature-aided tracking, target classification, and target fingerprinting to monitor and characterize each vehicle in the surveillance region. This set of tracks will supply the data from which behavioral features will be extracted. These long-term tracks and the behavioral features will feed a threat detection algorithm that will detect abnormal behaviors. In Phase I, Toyon will perform a feasibility study of the proposed solution by developing and extending component data fusion algorithms, designing the behavioral feature models, and evaluating anomaly detection performance to meet the specified technical objectives of the project.

User Systems, Incorporated
2137 Defense Highway, Suite 12
Crofton, MD 21114
Phone:
PI:
Topic#:
(410) 451-6799
Kancham Chotoo
OSD12-LD1      Awarded: 2/25/2013
Title:Autonomous Sensing and Deciding Framework Processor
Abstract:The goal of this SBIR effort is to develop an autonomous capability that image analysts will utilize to extract useful information from a large volume of data in real time. A key component of this capability is the development of a tool that will fully utilize very large intelligence data streams, e.g., those that collect data over spatially wide areas. Areas of interest (unusual activity) will be identified 10 to 100 times faster than analysts can do today. Further, the tool will discover key areas of interest that previously may have escaped identification. Our approach will focus on an innovative Unusual Activity or Inactivity Detector (UAID) using change detection products over wide areas and long periods of time that are commonly obtained from a synthetic aperture radar system. The changes include moving targets, arrival/departure of vehicles, tracks, etc. The UAID will ingest the high-resolution SAR data products and highlight only those areas/activities of high interest to image analysts and decision makers.

Applied Systems Intelligence, Inc.
3650 Brookside Parkway Suite 400
Alpharetta, GA 30022
Phone:
PI:
Topic#:
(678) 665-5668
John Merrihew
OSD12-LD2      Awarded: 5/23/2013
Title:Fusing Uncertain and Heterogeneous Information – Making Sense of the Battlefield
Abstract:ASI proposes to develop and evaluate a Phase I Demonstration System for the Improved Fusion Algorithm System (IFAS) based on its artificial intelligence technology for building Associate Systems. This Demonstration System will be based on requirements analysis with a written requirements document and will be evaluated by military operator users. Requirements analysis will include performance of cognitive task analysis to identify and detail scenarios which can be used to structure the Phase I demonstration system. The existing ASI Solomon (registered trademark) technology for Associate Systems will be augmented with stochastic networks using CVAR distributions and statistics according to the requirements. The requirements will also be used to structure user evaluations.

Charles River Analytics Inc.
625 Mount Auburn Street
Cambridge, MA 02138
Phone:
PI:
Topic#:
(617) 491-3474
Erik Thomsen
OSD12-LD2      Awarded: 5/13/2013
Title:A Decision Support Tool using Canonical Representations and Algorithms for Data Fusion and Information Delivery (CRA-FIND)
Abstract:Members of the armed forces depend on the quality of their information for mission success. Providing high quality information requires optimizing the value of the dynamic “information economy,” where data fusion is the supply, information requirements are the demand, and information delivery is how they connect. Though data fusion, requirements capture, and information delivery are often framed as algorithm problems, lessons from industry show that information quality problems are also caused by the lack of standard mathematical models (i.e., canonical forms) for multi-level representations of information. Canonical representations link data fusion, requirements capture, and information delivery through a mathematically sound, data- and model-driven approach. We therefore propose to design and demonstrate a decision support tool using Canonical Representations and Algorithms for Data Fusion and Information Delivery (CRA-FIND) that provides a: (1) test environment for data fusion and information delivery algorithms; (2) software-instantiated method that links algorithms via canonical representations to a shareable world model, thereby improving their robustness and precision; (3) systematic method for learning the contexts under which different algorithms perform well and using this information to manage those algorithms’ use; (4) general, reusable algorithm that queries the world model for information that can impact an algorithm’s outputs.

Toyon Research Corp.
6800 Cortona Drive
Goleta, CA 93117
Phone:
PI:
Topic#:
(805) 968-6787
Charlene S. Ahn
OSD12-LD2      Awarded: 5/23/2013
Title:Fusing Uncertain and Heterogeneous Information – Making Sense of the Battlefield
Abstract:Today's decision analysts are faced with an ever-increasing amount of information from a large number of sensors and other data sets. Data fusion is a key component of this effort to integrate information from various data sources. In particular, one of the crucial aspects in battlefield operations concerns the development of continuous, unambiguous tracks on objects of interest in the surveillance area. Toyon Research Corporation proposes to research and design a system that integrates and fuses data from disparate sensors in order to aid command-level decisions, particularly by persistently tracking targets of interest. Toyon will develop stochastic methods for identifying decision points in the problem of integrating information through data fusion. In addition, advanced inference algorithms to integrate information will be developed and enhanced, based on Toyon’s previous work with its Fusion and Correlation for Tracked Object Retention (FACTOR) module, which has submodules for kinematic fusion and for inferencing algorithms.

Aptima, Inc.
12 Gill Street Suite 1400
Woburn, MA 01801
Phone:
PI:
Topic#:
(781) 496-2465
Charlotte Shabarekh
OSD12-LD3      Awarded: 4/4/2013
Title:TERRAIN: Temporal Exploitation and Reasoning using Resource-Activity Inference Networks
Abstract:Detecting targets and predicting threat networks hidden in high volume all-source intelligence is a central challenge in the intelligence Processing, Exploitation and Dissemination (PED) cycle. Failure to maintain a common picture and shared situational awareness across disparate sensor products can result in enormous information loss, placing the entire team and intelligence community at a severe disadvantage. By improving the tasking of sensors, only the most salient intelligence will be collected, thus reducing the volume of data to be analyzed and optimizing the PED cycle. Aptima proposes to develop Temporal Exploitation and Reasoning using Resource-Activity Inference Networks (TERRAIN), an integrated system for Tasking, Collection, Processing, Exploitation and Dissemination (TC-PED). TERRAIN system will innovatively combine three critical functions in support of Marine Corps missions: active sensor stream analysis, threat situation estimation, and sensor allocation planning to produce a sensor tasking system that maximizes limited resources by identifying gaps in the current coverage and forecasting where future sensor coverage need will be greatest. Designed with Hadoop’s Map-Reduce Architecture for integration to a cloud-based computing environment, TERRAIN will be a lightweight, online system that scales to a high volume of streaming, multi-intelligence sensor feeds.

Commonwealth Computer Research, Inc.
1422 Sachem Pl., Unit #1
Charlottesville, VA 22901
Phone:
PI:
Topic#:
(434) 299-0090
Kevin Corby
OSD12-LD3      Awarded: 4/4/2013
Title:Intelligence Driven Intelligence Collection
Abstract:As the number and type of sensors and associated platforms grows, the complexity of collection management skyrockets. Achieving the goal of developing more effective and efficient collection plans requires a more informed collection management information architecture. This architecture must not only optimize collection plans based on infor-mation requirements and priority, but it must leverage the data that already exist within the intelligence repositories. These data can be used in two ways. First, intelligence re- positories can be scanned to determine if information requirements can be satisfied using the data already collected instead of creating more collection requirements. Second, when collections are necessary, they can be done more effectively by leveraging the power of predictive analytics to anticipate where the targets will be. In this proposal, CCRi describes a three phase collection management architecture that will guide users to creating unambiguous, complete, and well structured collection re- quirements that are both machine interpretable as well as human readable. The system will seek opportunities to avoid collection using existing data, and then optimize collec-tions using state-of-the art optimization techniques in conjunction with predictive analytic outputs.

Intelligent Automation, Inc.
15400 Calhoun Drive Suite 400
Rockville, MD 20855
Phone:
PI:
Topic#:
(301) 294-4241
Onur Savas
OSD12-LD3      Awarded: 4/3/2013
Title:Data-Centric Sensor Planning in a Cloud Environment
Abstract:The Department of Defense (DoD) has initiated a Data-to-Decisions (D2D) program to develop an open-source architecture system that enables rapid integration of existing and future data exploitation tools to achieve a new paradigm in the management and analysis of data. Big Data solutions, e.g., cloud computing and Hadoop/MapReduce, has shown their advantages on Big Data storage and intelligent analytics in many commercial applications, and is considered to be adopted in the D2D program. One of the applications that could benefit is automated sensor planning (or management) based on shared situation awareness (SA). In other words, sensors can be dynamically tasked (or re-tasked) based on the latest status of information requirements and on-line analytic predictive processing (OLAP). To address this need, we propose a systematic approach based on cloud computing that provide scalable data mining/analysis algorithms as well as tools and platforms for ingesting real- time sensor data (e.g., technical, semantic, unstructured) for shared situational awareness and predictive processing, and drive the sensor planning loop. The proposed product will be extensively tested in a cloud computing environment.

Lakota Technical Solutions, Inc.
PO Box 2309
Columbia, MD 21045
Phone:
PI:
Topic#:
(410) 381-9780
William J.Farrell
OSD12-LD3      Awarded: 4/2/2013
Title:Data to Decisions, Information Systems Technology
Abstract:Under this SBIR program, Lakota Technical Solutions, Inc. (Lakota) will develop a Level 4 fusion (Process Refinement) capability implemented within a cloud computing architecture. Leveraging an existing Multi-Sensor Exploitation Management System (MSEMS), Lakota will define cloud-based services for: (1) multi-INT sensor information storage and retrieval, (2) automated information needs assessment, and (3) assignment of sensor tasks to sensor resources. To realize a scalable capability with respect to the number of sensor assets and information needs, open-source cloud computing technologies will be employed to execute and distribute the MSESM functionality across a network of commodity hardware using a distributed semantic knowledge base for efficient storage, retrieval, and reasoning about multi-INT sensor information. Using the modular, extensible, and maintainable MSEMS architecture in conjunction with the scalability of cloud computing, the proposed approach offers a large-scale process refinement framework that aligns sensor collection needs with changing mission requirements and knowledge fidelity.

Charles River Analytics Inc.
625 Mount Auburn Street
Cambridge, MA 02138
Phone:
PI:
Topic#:
(617) 491-3474
Michael Farry
OSD12-LD4      Awarded: 5/21/2013
Title:A system for Using Knowledge of Narratives for Optimizing Workspaces to Enhance Reasoning (KNOWER)
Abstract:The wealth of data provided by modern information fusion (IF) tools to intelligence analysts exceeds their ability to effectively process, exploit, and disseminate actionable intelligence. This issue arises because these IF tools define information value and usage statically, and consider analysts rational and deterministic system components, rather than dynamic, critically thinking individuals who may exhibit biases and cognitive overload. To overcome those issues, we propose to design a novel IF system for Using Knowledge of Narratives for Optimizing Workspaces to Enhance Reasoning (KNOWER). Our solution consists of three main thrusts. First, we will develop a framework for using narrative within IF to provide a more naturalistic, intuitive means for framing human consumption and communication of information. Second, we will use that framework in the design of a workspace that provides intuitive interfaces for composing, reasoning about, and refining analytical hypotheses derived from IF in terms of narratives. Third, we will design IF methods that can respond to analyst information needs, expressed in terms of narrative elements. We will design and demonstrate KNOWER to establish the feasibility of our approach, and to establish an evaluation plan and performance metrics for Phase II.

Soar Technology, Inc.
3600 Green Court Suite 600
Ann Arbor, MI 48105
Phone:
PI:
Topic#:
(734) 887-7621
Jack Zaientz
OSD12-LD4      Awarded: 8/7/2013
Title:Intuitive Information Fusion and Visualization
Abstract:Traditional Information Fusions (IF) systems need to be expanded to encompass a decision support system (DSS) role that enables analysts and commander to manage the volume and heterogeneity of sensor data, including the textual ‘soft’ data descriptive of the social cultural landscape characteristics of contemporary missions. To achieve these improvements, SoarTech, supported by NCSU, will develop Sensemaking Patterns for Analysis and Decision-making (SPADE) a hybrid human computer interaction IF/DSS systems that draws its requirements from abductive hypothesis generation, a feature of sophisticated human decision models including sense-making and recognition primed decision-making. SPADE will apply hierarchical task network and hypothesis graphs to support temporal, spatial, semantic hypothesis definitions. SPADE uses user populated hypothesis patterns to auto- generate and adapt queries for existing, vetted data processing algorithms such as sentiment analysis and text indexing. This will enable analysts and decision makers to easily encode analytic and mission-planning hypothesis, including decision and time constraints, and automatically gain data management and integration, emergent concept detection, and hypothesis testing across traditional INTs and social data. SPADE will better support the coupling of data and decision needs, better mitigate analysis biases, and better support analysis and decision-making under uncertainty and time constraints.

DECISIVE ANALYTICS Corporation
1235 South Clark Street Suite 400
Arlington, VA 22202
Phone:
PI:
Topic#:
(703) 414-5032
Timothy Hawes
OSD12-LD5      Awarded: 2/27/2013
Title:Event Attribute Recognition and Labeling (EARL)
Abstract:Today’s intelligence analysts are overwhelmed with textual data in the form of Human Intelligence and open source data from the web. The amount of texts that analysts have access to is far greater than one could ever read. This is a fundamental problem for analysts who have pressing strategic and tactical deadlines. Innovations in event extraction help resolve this problem by turning unstructured text in to structured data stores of events. However, with millions of events in a database, simple event extraction does not sufficiently contribute to analysts’ Situational Awareness (SA). To truly increase SA, events must be searchable based on how, when, and if they occurred. This requires the ability to automatically recognize event attributes with very high accuracy. Under the Event Attribute Recognition and Labeling (EARL) effort numerous innovations towards high quality event attribute extraction are made. The EARL approach will use a multi-task classifier that simultaneously labels all attributes, with better accuracy than existing approaches. The EARL approach exploits deep linguistic features extracted from unstructured text. Finally, the EARL approach radically augments the data available, inexpensively and efficiently, by using crowdsourcing. The result is a capability that will far exceeded the current state-of-the-art in event attribute recognition.

Intelligent Automation, Inc.
15400 Calhoun Drive Suite 400
Rockville, MD 20855
Phone:
PI:
Topic#:
(301) 294-5214
Kaizhi Tang
OSD12-LD5      Awarded: 3/1/2013
Title:LASER: a Linguistic enriched And Scalable Event attribute extRaction System
Abstract:With the fast growth of web data and HUMINT reports, intelligent analysts need the capability to rapidly monitor and analyze event information in those massive amounts of unstructured textual data, in order to achieve and maintain persistent Situational Awareness (SA). Intelligent Automation, Inc., along with our collaborators, proposes to develop a Linguistic enriched And Scalable Event attribute extRaction System: LASER. There are three major innovations in LASER, firstly, it integrates even richer and more specialized features into each of the four classification models for the four event attributes, namely, modality, polarity, genericity and tense; secondly, it adopts robust classification models that can handle unbalanced class problem (which is the case in event attribute extraction); thirdly, it incorporates three post-correction approaches which are expected to bring more performance gains to event attribute extraction. LASER also leverages a state-of-the-art event extraction system which extracts relative high quality events such that event attributes can be further extracted. Furthermore, LASER uses powerful cloud computing architecture for information management and algorithmic computation.

Language Computer Corporation
2435 N. Central Expressway Suite 1200
Richardson, TX 75080
Phone:
PI:
Topic#:
(972) 231-0052
Sean Monahan
OSD12-LD5      Awarded: 2/27/2013
Title:Recognizing Event Attributes in Unstructured Text (REACT)
Abstract:In Phase I of REACT, we will demonstrate how extraction of attributes dealing with modality, polarity, and genericity can enhance the quality of information provided by a state-of-the-art event extraction and coreference system to provide actionable intelligence to analysts. We will demonstrate how our research makes significant improvements to the understanding of polarity and veridicality, the characterization of events as generic or episodic, the inference of author perspectives, and the fusion of event attributes across mentions to enhance the knowledge surrounding and improve the analyst’s situational awareness. We plan to leverage existing state-of-the-art natural language understanding and content extraction capabilities including (1) wide coverage event recognition and coreference resolution, (2) open-domain, customizable information extraction, and (3) methods for determining the social intentions of authors in text. This will enable us to extract events and the rich forms of semantic and pragmatic information expressed in their attributes in order to find and fuse information that satisfies the demands of today’s information analysts.

DECISIVE ANALYTICS Corporation
1235 South Clark Street Suite 400
Arlington, VA 22202
Phone:
PI:
Topic#:
(703) 414-5139
Mark Frymire
OSD12-LD6      Awarded: 4/3/2013
Title:AUDIO-based Cloud-enabled Language and Intelligence Processing (AUDIO-CLIP)
Abstract:Military operational tempo requires rapid development of highly accurate intelligence products from massive stores of data. Manual processing of data on this scale is infeasible. Automated processing systems have been developed to help analysts find the most relevant data to their current information requirements. This automated processing allows analysts to focus their efforts on performing higher level analysis. Unfortunately, most of these tools require the data to be in proper English. However, a significant percentage of the most fruitful data is in the native language of the areas of operation. Due to high error rates of automated transcription and translation tools, the majority of analytical tools are unable to produce meaningful results from this data. Therefore, DAC proposes to the AUDIO-based Cloud-enabled Language and Intelligence Processing (AUDIO-CLIP) system. The AUDIO-CLIP system will be focused on making all collected audio data available for real-time intelligence analysis and product generation. The algorithms within the system will focus on improving the automated generation of intelligence from foreign audio. AUDIO-CLIP will operate on foreign audio, transcriptions, and translations with the ultimate goal of improving the accuracy of the English transcripts for use by downstream analytical engines while also directly extracting intelligence at each step.

Li Creative Technologies
25 B Hanover Road, Suite 140
Florham Park, NJ 07932
Phone:
PI:
Topic#:
(973) 822-0048
Qi (Peter) Li
OSD12-LD6      Awarded: 4/5/2013
Title:Text Analytics from Audio
Abstract:We propose to design and implement a system that combines multiple audio transcription and multiple translation tools with natural language processing capabilities, such that information can be automatically extracted from audio files with improved performances. In our approach, speech waveforms of a foreign language are first processed to remove background noise using our noise reduction algorithm developed based on our patented auditory transform theory. The clean speech is then feed forward to multiple speech recognizers to convert to text. A error correction algorithm is then applied to reduce word error rates based on the text and a neural language model. Following that, multiple machine translators are used to translate the text to English. An error correction algorithm with English language model is then applied to make further correction. Finally, a natural language processing unit extracts out the entities, associations, concepts, and themes. Based on recent research results, the proposed approach has the potential to reduce word error rates by 20% and improve the entire system robustness. Our solution will leverage our experience and expertise in noise reduction, speech enhancement, neural network training, robust speech recognition and language model construction.

Progeny Systems Corporation
9500 Innovation Drive
Manassas, VA 20110
Phone:
PI:
Topic#:
(703) 368-6107
Gary Sikora
OSD12-LD6      Awarded: 4/4/2013
Title:Text Analytics from Audio
Abstract:Audio transcription, translation and Natural Language Processing (NLP) capabilities are required to automate the extraction of actionable information from foreign language audio files. The problem is that these language processing components are traditionally stand- alone and pipelined together producing accumulative growing word error rates, significantly degrading the level of trust. To overcome the shortfalls of a multi-staged pipelined approach we propose a two stage approach comprised of Automated Speech Recognition (ASR) transcription, then a Foreign Language Analysis (FLA) stage that goes directly from foreign language text to English concept encoded in FrameNet semantic frames. Given the innovation and newness of the FLA approach our proposal focuses on this stage, leveraging internal stages of SYSTRAN’s translation capabilities – as commercial device technologies evolve, more and better ASR solutions will be inherent available such as the iPhone Siri and Android Google Voice. The ability to have a single component or app to go directly from these device ASRs to semantic frames will result high levels of trust while easing integration with applications that need to consume actionable information. The plan is to produce 200 Farsi sentences with various tones and noise, use a commercial Farsi ASR and co-develop a Farsi FLA with SYSTRAN.

Arctan, Inc.
2200 Wilson Blvd. Ste. 102-150
Arlington, VA 22201
Phone:
PI:
Topic#:
(202) 379-4723
Michael Morefield
OSD12-LD7      Awarded: 5/2/2013
Title:Tactical Information Management (OCCAM)
Abstract:A system to automatically identify mission-relevant information within modern tactical databases, score them relative to current mission state, and transmit the most important elements in a timely manner to small unit leaders and other decision elements – subject to the constraints of both human and digital tactical bandwidth.

Broadata Communications, Inc.
2545 W. 237th Street, Suite K
Torrance, CA 90505
Phone:
PI:
Topic#:
(310) 530-1416
Prachee Sharma
OSD12-LD7      Awarded: 5/2/2013
Title:Broadata Tactical Information Management System
Abstract:The amount of data streaming into tactical systems is overwhelming the operators. The amount of information is so much that the situational awareness regarding the unfolding situations is difficult to extract from large amounts of data. The goal of this SBIR is to design and develop a framework for managing information with emphasis on the relative value of that information for a wide range of operations. To meet the goals of this SBIR, we propose a Tactical Information Management System or TIMS in this proposal. TIMS can be installed at the combat operations center. In this case, the data will be available from multiple sources and stored in the databases. TIMS shall implement detailed algorithms for filtering and prioritizing the data. The filtered data will be forwarded to the warfighters over low throughput links or to higher or adjacent command as need. Data filtering and prioritization algorithms in TIMS will include data processing and condition algorithms, algorithms to perform indexing and semantic analysis of the algorithms. A detailed approach for ranking the data is described. Multiple data attributes will be considered in information prioritization.

Intelligent Models, Inc.
9710 Traville Gateway Drive
Rockville, MD 20850
Phone:
PI:
Topic#:
(240) 401-9746
Yuri Levchuk
OSD12-LD7      Awarded: 5/2/2013
Title:Dynamic Information Assessment and Management Of Network Dissemination(DIAMOND)
Abstract:Recent revolutions in sensor technologies, coupled with a worldwide cultural phenomenon of social networking, offer an unprecedented window into the day-to-day lifecycles of human societies and regional populations. At the same time, the current trend toward higher- resolution, larger fields of view of existing and planned ISR systems is driving data volumes exponentially, exacerbated by the continuous demands for extending what can be inferred from data. This escalated data volume is overwhelming the cognitive constraints experienced by the intelligence analysts and turning the latter into a crucial bottleneck that hinders the overall system effectiveness. The negative impacts percolate downstream to the distributed warfighters operating tactical computing devices and upstream to the Combat Operations Centers. To promote efficient discovery of mission insights and to streamline their communication and processing by human operators, we propose DIAMOND (Dynamic Information Assessment and Management Of Network Dissemination) -- a novel intelligent automation engine equipped with scalable computing and visualization algorithms to automate the value-of- information assessment and concomitant value-based filtering, triage, fusion, and intuitive visualization of mission-critical Intel feeds. DIAMOND will help users to quickly discover, prioritize and highlight relevant information, and effectively communicate valuable Intel that elucidates relevant (but potentially dispersed and obscured) information.

Mercury Data Systems
4214 Beechwood Drive Suite 105
Greensboro, NC 27410
Phone:
PI:
Topic#:
(336) 294-2828
John Taylor
OSD12-LD7      Awarded: 5/2/2013
Title:Tactical Information Management
Abstract:We will develop a modular, distributed Bayesian inference system to provide a networked context inference system capable of operating on processor constrained devices. The Bayesian inference system will enable task automation via capabilities to extract relevance of situational awareness information, functional processes to develop plans and policies and to determine the utility of plan/policy options. The small unit leader will also be supported by a Decision-Centric user interface.

Boston Fusion Corp.
1 Van de Graaff Drive Suite 107
Burlington, MA 01803
Phone:
PI:
Topic#:
(617) 583-5730
Connie Fournelle
OSD12-LD8      Awarded: 4/11/2013
Title:Semantic Targeting using Analyst Role, Topics and Entity Recognition (STARTER)
Abstract:Typical search technology ignores the intended usage of documents—missing the opportunity to tailor prioritization and anticipate supplemental materials. Semantic targeting—tailoring content to the user’s predicted interests or needs—shows promise for improving analytical workflows. What is needed is a sophisticated approach to incorporating the user as part of the data model, leveraging the data, behaviors and interests of the user, and of prior analysts. Because analyst turnover is rampant, embedding legacy knowledge into the data model is critical for capturing guidance of trained analysts and helping new analysts conquer steep learning curves. We will assess the state-of-the-art, develop a system concept, and build a prototype to automatically adapt search using analysts’ roles. We will learn how role-specific individuals select and navigate content, follow leads, and identify relevant data in complex, multi-step tasking to accomplish this vision. To minimize the training data burden, we rely on minimally- invasive approaches that discover the roles, semantic content, and semantic connections. We will leverage and extend work in probabilistic topic modeling to include the role information and complex operational tasking. We will leverage our experience on the ONR-sponsored Mission-Focused Autonomy program to develop a plan to transition STARTER to an intelligence analysis environment.

Next Century Corporation
7075 Samuel Morse Drive Suite 250
Columbia, MD 21046
Phone:
PI:
Topic#:
(443) 545-3175
Todd Hughes
OSD12-LD8      Awarded: 4/16/2013
Title:Semantic Targeting for Open Source Intelligence Analysis
Abstract:Next Century Corporation proposes to develop text analytic technology that crosses the semantic gap at the shallow (but broad) area of event representation. The Event Representation and Structuring of Text (EVEREST) system will search for mappings to a semantic event model, interactively suggesting evidence for the occurrence of whole or partial events for human analysis and reporting. Our semantic targeting approach extends the ideas Open Information Extraction, Event Web, Semantic Web, and the Ozone Widget Framework. We believe that an event-centric approach will be critical for generating narratives that confer meaning upon large, complex, uncertain, and incomplete data sets.

Smart Information Flow Technologies, d/b/a SIFT
211 N 1st St. Suite 300
Minneapolis, MN 55401
Phone:
PI:
Topic#:
(781) 799-3603
Mark Burstein
OSD12-LD8      Awarded: 4/10/2013
Title:STRIDER: Semantic Targeting of Relevant Individuals, Dispositions, Events, and Relations
Abstract:The proposed STRIDER (Semantic Targeting of Relevant Individuals, Dispositions, Events, and Relations) technology combines an open source semantic parser with task-specific knowledge structures and diagram-based user interfaces to automatically extract, display, and record mission-relevant information for intelligence analysts. The approach works by (1) gathering details about the analyst's objectives using a link diagram interface, (2) performing semantic analysis to extract objective-relevant information from a large corpus, and (3) displaying and recording the extracted information for the analyst. STRIDER integrates with existing link diagram products to consume and enrich the information the analyst has gathered and recorded through other means. This ultimately reduces the time and cognitive load of finding, analyzing, and recording information for intelligence analysis problems. STRIDER is semi-automatic information extraction technology that is guided by the user's objectives and the user's attention.