Research Summaries

Back Provisioning of a multi-int fusion environment in the era of AI and Machine Learning

Fiscal Year 2019
Division Research & Sponsored Programs
Department NPS Naval Research Program
Investigator(s) Godin, Arkady A.
Sponsor NPS Naval Research Program (Navy)
Summary The era of AI and Machine Learning / Deep Learning (ML/DL) has a significant effect on TCPED. Dissemination architecture should support an easy integration providing decision makers (at any of the tiers including the tactical edge with its D-DIL connectivity constraints) to do more than visual exposure to the valuable information. The new era dissemination goal is to enable decision-makers with AI and ML/DL capabilities in decentralized environment. For instance, METOC SME should have tools capable to fuse organic hard-to-access METOC data sources in-situ without a need to disseminate such data to the higher tier for integration/fusion AI reasoning and ML/DL training model & analysis.

Research Methodology/Plan will:
• Identify distributed storage for canonical representation of all post-interpretation data modalities; NPS research views sparse-dense storage as right for integration/fusion cell for dissemination
• Design ontological knowledge representation facilitating AI-based integration and multi-int fusion
• Explore integration and multi-int fusion approach based on canonical array-based distributed storage
• Identify distributed computing framework to execute AI-based multi-int fusion
• Propose dissemination approach to concurrently support: client-based visualization/navigation and client reconstruction of AI reasoning environment from client array storage
• Propose an approach to transition from Forecaster-In-The-Loop to Forecaster-On-The-Loop decisions where automation of mission-specific METOC guidance used in decision analytics is accomplished via AI and ML/DL techniques

Deliverables include: report on distributed storage with canonical capabilities to model data necessary for multi-int fusion; describe the role of ontologies, if any, for multi-int fusion and multi-int fusion provisioning; report on AI/ML/DL computational frameworks; describe proposed dissemination approach which enables AI reasoning on remote devices.
Keywords
Publications Publications, theses (not shown) and data repositories will be added to the portal record when information is available in FAIRS and brought back to the portal
Data Publications, theses (not shown) and data repositories will be added to the portal record when information is available in FAIRS and brought back to the portal