Autonomous Multi-Perspective Hierarchical Centralized Fault Monitoring System

Singularity-Intelligence Amplified, LLC (Singularity-IA) proposes the development of a multi-perspective hierarchical Centralized Fault Monitoring System (CFMS).  We plan to pursue a Modular Open Systems Approach (MOSA) by dividing the CFMS functionality into three portions: data import, data storage, and visualization.  We plan to develop each of these functions as separate modules, allowing them to plug into other future systems. Crew members have different training and roles to perform on a ship like the Constellation Class Frigate.  Therefore, our approach to visualization recognizes that different roles might require different methods for visualizing the current situation.  We identify these perspectives as: functional, regional, and system-level.  Essentially, this allows crew members to be presented a perspective and receive information structured within the context of their role on the ship. This accelerates their ability to absorb information. Modeling and simulation, using the Modelica modeling language, allows us to leverage our expertise to develop models of the ship and its important systems, subsystems, and components.  We can use these models to seed fault conditions and run simulations to generate data resembling current data acquisition systems that the ship may use.  This allows us to create a significant number of different fault scenarios, of varying levels of complexity, to provide a strong foundation for testing and analysis of our proposed solution. Adding the modeling and simulation capabilities allows us to leverage some of our existing technologies in the Fault Management (FM) space, which provides an interesting path forward.  Instead of just monitoring fault conditions, we can leverage the data acquired by a monitoring system, process it with a virtual cognitive ability using digital twin models, and leverage corrective action abilities to support autonomous real-time active fault management. The phase I project goal is to demonstrate the feasibility of this approach and tease out the requirements for future development and implementation of a Centralized Fault Monitoring capability that incorporates advancements in human & machine cognitive interfacing to yield faster and more effective fault management in adverse, multi-failure conditions.