A Fault Detection and Isolation Method Applied to Liquid Oxygen Loading for the Space Shuttle
Abstract
Process-monitoring and fault location techniques have been developed at the Kennedy Space Center in a domain of mixed media control in NASA's Space Shuttle Launch Processing System. An intuitively appealing diagnostic technique and representation of the system's structure and function were formulated in cooperation with system engineers. Functional relationships that determine the consistency of sensor measurements are represented by symbolic expressions embedded in frames. Functional relationships are stored in exactly one place, so they must be inverted to determine hypothetical values for possibly faulty objects. Propagating these hypothetical states to other sensors permits the location of faults. Standard symbolic inversion techniques have been extended to include conditional functions. A demonstration system is operating, and its evaluation will soon use live data from the firing rooms at KSC.
Cite
Text
Scarl et al. "A Fault Detection and Isolation Method Applied to Liquid Oxygen Loading for the Space Shuttle." International Joint Conference on Artificial Intelligence, 1985.Markdown
[Scarl et al. "A Fault Detection and Isolation Method Applied to Liquid Oxygen Loading for the Space Shuttle." International Joint Conference on Artificial Intelligence, 1985.](https://mlanthology.org/ijcai/1985/scarl1985ijcai-fault/)BibTeX
@inproceedings{scarl1985ijcai-fault,
title = {{A Fault Detection and Isolation Method Applied to Liquid Oxygen Loading for the Space Shuttle}},
author = {Scarl, Ethan A. and Jamieson, John R. and Delaune, Carl I.},
booktitle = {International Joint Conference on Artificial Intelligence},
year = {1985},
pages = {414-416},
url = {https://mlanthology.org/ijcai/1985/scarl1985ijcai-fault/}
}