Ph.D. Dissertation Proposal

by

Peter Zane Schulte

Advisor: Dr. David Spencer

A State Machine architecture

for aerospace vehicle fault protection

12:30 PM, Monday, January 30, 2017

Montgomery Knight Building

Room 317

ABSTRACT:

Because of their complexity and the unforgiving environment in which they operate, aerospace vehicles are vulnerable to mission-critical failures. In order to prevent these failures, aerospace vehicles often employ Fault Detection, Isolation, and Recovery (FDIR) systems to detect, identify the source of, and recover from faults. Typically, aerospace systems use a rule-based paradigm for FDIR where telemetry values are monitored against specific logical statements such as static upper and lower limits. The model-based paradigm allows more complex decision logic to be used for FDIR. State machines are a particular tool for model-based FDIR that have been explored by industry but not yet widely adopted. This study develops a generic and modular state machine FDIR architecture that is portable to flight software. The study will focus on FDIR for the Guidance, Navigation, & Control subsystem, but it will be presented in a manner that is applicable to all vehicle subsystems. This architecture applies to a wide variety of missions and vehicles and contains components that can be rearranged, added, or removed easily. The architecture is developed in a way that is straightforward to export to flight software via autocoding. Two specific case studies are employed to demonstrate the architecture. The first is a terrestrial application of unmanned aerial vehicles for 3D scanning and mapping, which is validated through flight testing. The second is a space-based application of automated close approach and capture for a Mars sample return mission, which is validated through processor-in-the-loop testing with flight-like avionics components.

Dr. E. Glenn Lightsey

Aerospace Engineering

Georgia Institute of Technology

Dr. Mark Costello

Aerospace Engineering

Georgia Institute of Technology

Dr. Neil Smith

Visual Computing Center

King Abdullah University of Science and Technology

Mr. Paul Rosendall