N131-062 TITLE: Advancing the State of the Art in Artificial Intelligence for Simulation Training

TECHNOLOGY AREAS: Information Systems, Human Systems

OBJECTIVE: Develop Artificial Intelligence (AI) software that is generalized across entity-level simulation systems and can be used in training simulation to generate entity behaviors that are both contextually and tactically realistic.

DESCRIPTION: The United States Marine Corps (USMC) has been a pioneer in using commercial gaming technology for training. The current small unit training simulation is VBS2, developed by Bohemia Interactive Simulations. It is used primarily as a small unit training system in which each squad member has their own networked laptop and controls their own virtual character. A contracted role player usually controls opposing force characters, civilian characters, and friendly units. Although a human controller adds behavioral and tactical realism to virtual training scenarios, such interactions are costly which limit the number of Marines that can receive training. We are seeking solutions to develop AI capabilities to reduce the number of contracted role players required to supervise and manage simulation characters. In addition, other entity-level simulations, such as Joint Semi-Automated Force (JSAF), are used as part of the Deployable Virtual Training Environment. There is a desire to support improved aircraft Terminal Attack behaviors in support of future programs.

PHASE I: Define and develop a concept for using AI in VBS2 and JSAF to improve simulation training. The focus should be on improving the behavioral realism of the NPC while reducing the need for support personnel.

PHASE II: Produce prototype software based on Phase I work which can be set up and effectively used by minimally trained enlisted Marines.

PHASE III: The AI system may become an integrated part of the SITE Program of Record, supplementing and improving VBS2 simulation.

PRIVATE SECTOR COMMERCIAL POTENTIAL/DUAL-USE APPLICATIONS: The technology will have application in the commercial gaming arena, making games for entertainment more engaging and realistic.

REFERENCES:
1. Naval Research Advisory Committee. 2009. "Immersive Simulation for Marine Corps Small Unit Training." http://www.nrac.navy.mil/docs/2009_rpt_Immersive_Sim.pdf.

2. Swartout, William, Johnathan Gratch, Randall Hill, Eduard Hovy, Stacy Marsella, Jeff Rickel and David Traum. 2006. "Toward Virtual Humans." AI Mag. 27, 2, 96-108. http://www.isi.edu/natural-language/people/hovy/papers/05AI-mag-VirtualHuman.pdf.

3. White Paper: VBS2; Jan 06, 2012 http://products.bisimulations.com/VBS2_Whitepaper.pdf?q=whitepaper.

KEYWORDS: Simulation; modeling; artificial intelligence; training; VBS2; JSAF

** TOPIC AUTHOR (TPOC) **

DoD Notice:   Between 16 November, 2012 through 16 December 2012, you may talk directly with the Topic Authors (TPOC) to ask technical questions about the topics. Their contact information is listed above. For reasons of competitive fairness, direct communication between proposers and topic authors is not allowed starting 17 December, 2012, when DoD begins accepting proposals for this solicitation.

However, proposers may still submit written questions about solicitation topics through the DoD's SBIR/STTR Interactive Topic Information System (SITIS), in which the questioner and respondent remain anonymous and all questions and answers are posted electronically for general viewing until the solicitation closes. All proposers are advised to monitor SITIS (13.1 Q&A) during the solicitation period for questions and answers, and other significant information, relevant to the SBIR 13.1 topic under which they are proposing.

If you have general questions about DoD SBIR program, please contact the DoD SBIR Help Desk at (866) 724-7457 or email weblink.