Matthew J. Daigle
Lead, Diagnostics & Prognostics Group / Research Computer Scientist, Intelligent Systems Division, NASA Ames Research Center
NASA Ames Research Center
Real-time Safety Monitoring and Prediction for the National Airspace: Developed framework and algorithms for real-time safety monitoring and prediction of the national airspace. Demonstrated in simulation.
Prognostics and Decision-Making for Electric Aircraft: Researched and developed new battery models and algorithms for battery health monitoring with application to electric aircraft. Integration with mission planning and decision-making algorithms.
Physics-based Simulation of Cryogenic Propellant Loading Systems: Lead in development of physics-based simulation for cryogenic propellant loading system testbed. Developed physics models of components and testbed in Matlab/Simulink with fault injection capability for one- and two-phase flow. Supervised development of simulation architecture and user interfaces.
Model-based Prognostics with Application to Cryogenic Propellant Loading Systems: Researched model-based prognostics methods for pneumatic valves, current-pressure transducers, solenoid valves, centrifugal pumpbs, and lithium-ion batteries. Constructed pneumatic valve testbed for algorithm validation.
Battery Health Management for Spacecraft: Researched and developed new battery models and algorithms for battery health monitoring with application to spacecraft. Deployed for EFT-1 mission.
Prognostics and Decision-Making for Planetary Rovers: Researched and developed algorithms for battery health monitoring, integrated diagnosis and prognosis, future usage modeling, and uncertainty quantification. Developed physics-based rover simulation and framework for simulation-based algorithm validation.
Model-based Diagnosis with Application to Rapid Propellant Loading: Developed MATLAB/Simulink simulation of Rapid Propellant Loading testbed, including nominal and faulty operation. Developed simulation interface for fault injection, data recording, and integration with the Hybrid Diagnostic Engine (HyDE).
Vanderbilt University
Hybrid Systems Diagnosis with Application to Electrical Power Systems: Developed formal event-based framework for single and multiple fault diagnosis based on temporal orders of measurement deviations. Researched discrete fault diagnosis in hybrid systems based on qualitative algorithms. Implemented algorithms in the Fault Adaptive Control Technology (FACT) software. Applied to Advanced Diagnostics and Prognostics Testbed.
Distributed Diagnosis with Application to Mobile Robots: Researched distributed diagnosis of multi-robot systems. Implemented distributed fault detection and isolation algorithms and performed experimental studies for box-pushing and formation-keeping tasks.