SBIR/STTR Award attributes
Equipping the US military for operations in a rapidly changing world driven by data and technology is extremely difficult given the array of capabilities, threats, technologies, and objectives that are constantly changing. Improved methods for informing decision makers and determining what capabilities are effective against evolving threats are necessary to meet the needs of the warfighter in a timely manner. One effective method that accelerates analysis, concept exploration, and decision making is using Modeling and Simulation (M&S). The use of M&S to perform theater-wide mission analysis presents its own set of challenges. No one person, or team, is able to be aware of, mentally retain, or consider the full extent of the DoD’s simulation and modeling capabilities, let alone how they would integrate, leading to an inability to store, find, retrieve, and compose models for a particular need. Previous efforts to characterize, catalog, and compose existing model asset portfolios are inconsistent and prone to taxonomy incompatibilities, manual error, and interoperability issues. There is significant need for tooling that allows a team to represent existing models, catalogue and discover them, and evaluate them for suitability against simulation needs so they can be incorporated into the decision-making process. SimVentions is partnering with KaDSCi LLC and the Old Dominion University / Virginia Modeling and Simulation Center (ODU/VMASK) as Team SimV to address this problem. Team SimV’s prior work and domain expertise across these areas uniquely positions us to create a complete, practical, and useful architecture with workflows, automation, software tools, and data structures. We have demonstrated in our DP2 justification our readiness to apply the data structure approach and knowledge representation strategies for SMR across the aspects of security, representation, model correspondence, model suitability, integration, and management. By applying the tools and techniques of Machine Learning, Natural Language Processing, and Artificial Intelligence, our technical approach will enable the architecture needed to compose multi-mission, multi-domain, kill-web mission analysis in faster than real time. To realize the benefits, Team SimV proposes to build the SMR with four key innovations - a hybrid data storage approach consisting of both structured and unstructured data, a microservice-based classification solution decoupled from the SMR core, an iterative filtering interview workflow to identify relevant assets, and a robust scoring and composition technology. Team SimV’ innovative approaches will enable SMR to effectively identify and classify assets; create an architecture for addressing known types and to manage future unknown types; create a framework to evaluate interoperability and integration gaps; and give an environment to evaluate the level of effort and time to create solutions to theater-wide mission analysis questions.
