Aerospace vehicles are vulnerable to hardware and software faults that lead to mission-critical failures. Advances in onboard fault protection capability are necessary as both terrestrial and space vehicles increase in autonomy. State machines offer a useful tool for system behavior modeling and fault protection. This study presents an architecture for aerospace vehicle fault protection, focusing on the guidance, navigation, and control subsystem. The architecture is designed to be generic for use with any vehicle or mission; modular with components that can be added, removed, or rearranged; and portable for ease of conversion to flight software. A subsystem taxonomy delineates relevant vehicle hardware and software components. A fault tree analysis is performed to identify relevant faults. To model system mode behavior, a functional state machine is defined. A diagnostic state machine is developed for onboard model-based fault diagnosis. Finally, a system block diagram illustrates how fault and mode components can be integrated with other aspects of the system. Two specific case studies are presented, including an unmanned aerial vehicle application and a Mars sample return orbital rendezvous and capture scenario, demonstrating that the generic architecture can be adapted to diverse vehicles in very different regimes.
Schulte, P. Z., & Spencer, D. A. (2020). State Machine Fault Protection Architecture for Aerospace Vehicle Guidance, Navigation, and Control. Journal of Aerospace Information Systems, 17(2), 70–85. doi:10.2514/1.i010673
This material is based upon work supported by the National Science Foundation Graduate Research Fellowship Program under grant no. DGE-1148903. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the National Science Foundation. FalconViz and the King Abdullah University of Science and Technology (KAUST) provided funding, in-kind support, and technical guidance for development of the unmanned aerial vehicle nervous system. The Mars Sample Return fault protection work was completed under contract (Federal award no. 1568300, and subaward no. 4103-79588) with the NASA Jet Propulsion Laboratory (JPL) at the California Institute of Technology. McClain Goggin at Purdue University completed most of the Mars Sample Return trajectory design work. Neil Smith at KAUST provided ideas, mentorship, personal support, and technical guidance that were invaluable to this study. Special thanks to Rob Lock (Mars Program Office) and Peter Meakin (Fault Protection and Autonomy Group Supervisor) at the JPL for their mentorship and assistance. Also, the JPL Mars Sample Return study team and Rendezvous Working Group provided much input by participating in breakout discussions. Many engineers at the JPL were consulted to solicit ideas for fault protection research. Presented as Paper IAC-18-C1.5.11x45016 at the 69th International Astronautical Congress, Bremen, Germany, 01–05 October 2018.