US 12,488,225 B1
Modular open system architecture for common intelligence picture generation
Adam Estrada, Reston, VA (US); Dave Rabrun, Rockville, MD (US); Matthew Berra, Alexandria, VA (US); Kevin Naquin, Erie, CO (US); Michael Ludlam, Tampa, FL (US); Kristen E. Mistysyn, Erie, CO (US); Nicholas Stephens, Thornton, CO (US); and Jeffery Harlan, Ellicott City, MD (US)
Assigned to ROYCE GEOSPATIAL CONSULTANTS, INC., Arlington, VA (US)
Filed by Royce Geospatial Consultants, Inc., Arlington, VA (US)
Filed on Feb. 26, 2025, as Appl. No. 19/063,602.
Application 19/063,602 is a continuation in part of application No. 19/048,902, filed on Feb. 8, 2025.
Int. Cl. G06N 3/0475 (2023.01); G06N 3/063 (2023.01); G06N 3/092 (2023.01)
CPC G06N 3/0475 (2023.01) [G06N 3/063 (2013.01); G06N 3/092 (2023.01)] 14 Claims
OG exemplary drawing
 
1. A computing system for a modular open system architecture for common intelligence picture generation, the computing system comprising:
one or more hardware processors configured for:
receiving and processing priority intelligence requirements (PIRs) through a multi-intelligence fusion system large language model (LLM) configured to:
parse and analyze requirements to identify collection parameters and priorities;
generate courses of action for intelligence collection operations that maximize collection success rates while minimizing resource usage based on predefined operational constraints;
identify and rank collection assets and data sources based on requirement specifications;
maintain continuous learning from historical collection outcomes to select assets based on historical performance metrics including successful collection rate and data quality scores;
executing multi-phase intelligence processing comprising:
planning intelligence collection operations through LLM-enhanced automated requirement analysis and defining collection strategies that maximize target coverage while minimizing resource allocation;
collection of data through artificial intelligence (AI)-coordinated tasking of satellite, signal intelligence platforms, and open-source intelligence resources;
processing collected data through containerized analytics workflows;
analyzing the processed data through hybrid machine learning models;
combining computer vision and natural language processing;
dissemination of analyzed intelligence through context-aware intelligence product generation;
feedback through automated refinement mechanisms;
calculating collection feasibility by analyzing requirements against an integrated collection infrastructure including orbital coverage and revisit frequency;
identifying available collection assets by querying a multi-modal collection database comprising data about satellites, signal intelligence platforms, and open-source intelligence resources;
collecting environmental and operational data including space domain conditions, atmospheric conditions, and terrain factors;
developing an integrated collection plan through a containerized architecture configured for workflow generation, pattern analysis and dynamic exploitation;
processing collected intelligence through GPU-accelerated deep learning models for automated target recognition and pattern detection;
orchestrating multi-source collection by monitoring and assessing feasibility factors including environmental conditions, availability, and collection geometries;
maintaining domain awareness through integrated monitoring, environmental integration, asset management, and visualization;
fusing intelligence data across multiple subsystems to correlate findings across multiple intelligence disciplines in a common intelligence picture (CIP);
implementing automated workflows for intelligence analysis and dissemination while enforcing security controls and privacy protections; and
generating and updating intelligence products based on continuous collection and feedback integration.