Air force 14. 1 Small Business Innovation Research (sbir) Proposal Submission Instructions

PHASE I: Develop test and evaluation methodology for holographic and other light field 3D displays. Design an automated procedure for measuring the quality of FoLD systems. Deliverables shall include a review of literature and relevant technologies, a proposed strategy for optimizing the automated measurement procedure, a mature system design, and a draft handbook for FoLD test and evaluation.

PHASE II: Build and deliver the display measurement system designed in Phase I. Verify by demonstration that the measurement system is suitable for FoLD 3D visualization research, system calibration, and product acceptance testing. Obtain industry feedback and expand Phase draft handbook into a "Handbook FoLD Test & Evaluation Methodologies."

PHASE III DUAL USE APPLICATIONS: Develop a formal standard for FoLD test & evaluation methodology to enable transition and transfer of full-parallax 3D displays to military and civilian applications ranging from geospatial representations to modeling to design. Publish the standard via an international display metrology committee.

REFERENCES:

1. Hopper, D.G. et al., “Air Force Display Test & Evaluation Methodologies,” Draft Technical Report (March 2013). Available to U.S. Government Agencies and their Contractors.

AF141-023 TITLE: Voice-Enabled Agent for Realistic Integrated Combat Operations Training

OBJECTIVE: Develop and demonstrate a voice-enabled intelligent agent to improve the realism of integrated combat operations training and rehearsal within the Air Support Operations Center.

DESCRIPTION: In today’s asymmetric combat operations, the integration and interoperability of various mission areas of the Command and Control domain is growing. The state-of-the-art in training and deployment preparation for these constituent mission areas is to train the missions as separate stovepiped communities using live and simulated assets that are not consistently available. There are some limited opportunities for the various communities to interact with one another via a distributed network, but the availability of appropriate technologies, other players and timely scheduling of activities jeopardizes persistence in distributed training for these mission areas today. This topic is seeking a method to establish capabilities for each mission area to realistically and routinely train as if they were interoperating as they would do in theater. One of the most salient missing pieces to achieve realistic training is the lack of realistic voice-enabled agents that can regularly play the role of other trainees, wingmen, entities, and coordination agencies. The current state-of- the-art in linguistic modeling does not provide a mechanism for intelligent agents to communicate in a realistic manner using voice within simulation environments. Voice-enabled intelligent agents are typically developed in a closed environment that does not take systems like military simulators into account. The voice agent developed in this effort must be designed using industry standards to effectively interact within a simulation environment. During Phase I the voice agent should be a position that is either within the ASOC or that externally coordinates with the ASOC. This effort will develop a flexible voice-enabled agent technology to support training in and across the mission areas identified above. This effort will develop methods to capture real-world communication and coordination instances and develop software models for agents and teammates that can realistically communicate and coordinate training events as though human players were present. While we envision the developed agents from this effort to improve the realism of training in these mission areas directly, there is also a great potential that these agents can also provide realistic communication instances for an integrated Live, Virtual, and Constructive (LVC) deployment preparation and training capability for the constituent communities of practice.

PHASE I: This phase will identify content related to the ASOC internal or external operators for the development effort. In addition, Phase I will develop a rudimentary proof-of-concept desktop exemplar of the training and rehearsal concept to be fully developed in the Phase II effort. Develop preliminary transition plan.

PHASE II: Prioritize missions for scenario and content development. Evaluate scenarios in the environment focusing initially in the interaction between aircraft, ASOCs, and JTACs on the ground, then RPA coord and other air traffic control and coord entities as synthetic players. Evaluations will quantify training effectiveness and mission readiness enhancement resulting from the environment. Training transfer to live events and exercises will be assessed. Refine transition plan.

PHASE III DUAL USE APPLICATIONS: Uniquely capable and cost-effective training and rehearsal capability that can be included as a part of live and virtual training and rehearsal, which does not exist today for operational combat training. No similar approach to train multi-role and other manned/unmanned aircraft.

REFERENCES:

1. Ball, J. (2012). The Representation and Processing of Tense, Aspect & Voice across Verbal Elements in English. Proceedings of the 34th Annual Conference of the Cognitive Science Society. Sapporo, Japan: Cognitive Science Society.

AF141-024 TITLE: Adaptive Screen Materials for Image Projection

DESCRIPTION: In immersive training simulation environments, the primary stimuli presented to participants consists of visual imagery or cues. Some large-scale simulation environments are configured such that the imagery is front-projected on screens that (except for the floor) may partially or fully surround trainees. Some such environments are configured as large spherical or hemispherical domes in which the profile of the screen surface describes a compound curve. Such environments may be used for training both day and night operations. When training for day operations, a screen with high gain (e.g., high surface reflectivity), coupled with a set of projectors having high output, may be desirable for credible immersion. When training for night operations, participants may employ real night vision imaging devices (“NVGs”) to view night scenes projected on the screen, e.g., the NVGs are stimulated by the projected scene and produce intensified imagery. Ideally the intensified imagery seen in the simulator will credibly match the imagery that the NVGs would produce when used in a corresponding operational night environment. At the same time, the projected night scene must appear realistic to the unaided eye. NVGs are very sensitive, so when using screen materials with high gain, filters must be used or projector settings must be changed to greatly reduce scene brightness. Such approaches may result in color artificialities and/or projectors operating near the extreme lower end of their brightness gamut, causing limited remaining available dynamic range and imagery that appears unrealistic to the unaided eye. A front projection screen surface having very low gain for night environments would allow image projectors to operate at more normal settings, permitting greater dynamic range while also giving imagery appearing more realistic both through the NVG and to the unaided eye.

PHASE I: Research, define, compare and document technical capabilities and options. Determine a screen material concept capable of meeting all requirements in “Description” for immersive day, twilight and night simulation environments. Develop preliminary transition plan and business case analysis.

PHASE II: Demonstrate the proposed Phase I design concept with a prototype screen having an area of at least 3 square meters and that is scalable. Appearance of the imagery as viewed both with NVGs and by the unaided eye over a dynamic range spanning day to night is a key consideration. Submit a complete technical report documenting all work under the effort. Refine transition plan and business case analysis.

PHASE III DUAL USE APPLICATIONS: Military: Any training simulation system requiring realistic day, twilight and night environments for dismounted trainee participants. Examples: USAF JTAC Simulator; US Army Dismounted Soldier Simulator. Commercial: Entertainment and motion picture industry, also training or educational.

REFERENCES:

1. Joint Publication 3-09.3, Close Air Support, July 8, 2009, http://www.fas.org/irp/doddir/dod/jp3_09_3.pdf.

AF141-025 TITLE: Adaptive Instruction Authoring Tools

DESCRIPTION: The impact of intelligent tutoring systems and other forms of adaptive training technologies for promoting learning and subsequent performance has been shown a number of times in traditional classroom content areas such as reading and mathematics. Several applications of adaptive training have also been attempted with some success in more operational contexts such as electronics troubleshooting, power systems maintenance, and, most recently, in information technology troubleshooting and maintenance (see Fletcher, 1988; McCarthy, 2008). However, the development time associated with the creation of these tutors and adaptive training is significant and to date has involved a substantial amount of machine and knowledge engineering to achieve the desired end results. Moreover, high sustainment costs associated with retooling the content layers in these systems as operational domains change, as well as a lack of open source tools to facilitate growth of content databases and to promote domain expert generation of new content, continue to limit growth of these systems in the field. These issues, along with a persistent emphasis on closed and proprietary “one-off” tools, content, and architectures, has limited the broader application of intelligent tutoring and adaptive training systems in the military and especially in more complex operational domains. Prior work has made progress in the use of machine learning (cf. Stevens-Adams, et al., 2010) or high-order languages (cf. Cohen, et al., 2005; St. Amant, et al., 2005; Ritter, et al., 2006) to reduce development and sustainment costs. However, these approaches have characteristics that limit their suitability in a military training environment. For example, SMEs have to take significant time off task to learn the systems needed to create and update content, and there is still a significant dependence on proprietary methods that are costly and keep content from being more open and sharable across similar domains. This effort will address a number of these limitations by creating tools for authoring and maintaining adaptive training systems. These tools should provide intuitive and easy to use methods to allow domain SMEs and other instructional design and content developers with relatively little experience in the underpinnings of tutors or adaptive training system architectures to develop and sustain training applications that promote the benefits associated with this kind of instruction and training. Further, the tools should assist the developers in the application of best practices in instructional science and events of instruction such that the system is both content and instructionally valid. Finally, the developer tools must permit a more open and sharable design for content and instruction such that content and instructional approaches can be generalized across similar content spaces and training contexts.

PHASE I: Summarize best practices/applications in intelligent tutoring and adaptive training. Identify key common and unique features. Conduct a capabilities/gap analysis. Develop recommendations and a design specification for a tool set addressing the gaps while supporting more open and user friendly training design and delivery. Identify a content domain of relevance for a Phase II demonstration.

PHASE II: Develop and demonstrate the tool set in the content domain identified in Phase I. Implement the tools in an adaptive training exemplar and conduct user and training impact assessments. Refine tools and the exemplar, identify a domain to evaluate the generalizability and reuse of content and instructional approaches and conduct initial evaluations. Document impact of the exemplars on development and sustainment costs and on trainee performance.

PHASE III DUAL USE APPLICATIONS: The tools will improve training dev/delivery response times in military training squadrons. The tools permit a broader use in complex civilian areas such as emergency response and manned and unmanned border security and infrastructure monitoring where adaptive instruction is sorely needed.

REFERENCES:

1. Cohen, M. A., Ritter, F. E., & Haynes, S. R. (2005). Herbal: A high-level language and development environment for developing cognitive models in Soar. In Proceedings of the 14th Conference on Behavior Representation in Modeling and Simulation, 133-140.
2. Fletcher, J. D. (1988). Intelligent Training Systems in the Military. In S.J. Andriole & G.W. Hopple (Eds), Defense Applications of Artificial Intelligence: Progress and Prospects. Lexington, KY: Lexington Books.
3. McCarthy, J. E. (2008). Military Applications of Adaptive Training Technology. In M.D. Lytras, D. Gaševice, P. Ordóñez de Pablos, & W. Huang (Eds), Technology Enhanced Learning: Best Practices. Hershey, PA: IGI Publishing.
4. Ritter, F.E., Haynes, S. R., Cohen, M., Howes, A., John, B., Best, B., Lebiere, C., Jones, R. M., Crossman, J., Lewis, R. L., St. Amant, R., McBride, S.P., Urbas, L., Leuchter, S., & Vera, A. (2006). High-level behavior representation languages revisited. In Proceedings of the Seventh International Conference on Cognitive Modeling, 404-407. Trieste, Italy: Edizioni Goliandiche.
5. Stevens-Adams, S.M., Basilico, J. D., Abbott, R. G., Gieseler, C. J., and Forsythe, C. (2010). Using After-Action Review Based on Automated Performance Assessment to Enhance Training Effectiveness. HFES 2010. San Francisco, CA. 2010.
KEYWORDS: Tutoring, system, adaptive, training, open training solutions, shareable content, guided instructional design

AF141-026 TITLE: Distributed Mission Operations Gateway