Electrical power systems play a critical role in spacecraft and aircraft. This paper discusses our development of a diagnostic capability for an electrical power system testbed, ADAPT, using probabilistic techniques. In the context of ADAPT, we present two challenges, regarding modelling and real-time performance, often encountered in real-world diagnostic applications. To meet the modelling challenge, we discuss our novel high-level specification language which supports auto-generation of Bayesian networks. To meet the real-time challenge, we compile Bayesian networks into arithmetic circuits. Arithmetic circuits typically have small footprints and are optimized for the real-time avionics systems found in spacecraft and aircraft. Using our approach, we present how Bayesian networks with over 400 nodes are auto-generated and then compiled into arithmetic circuits. Using real-world data from ADAPT as well as simulated data, we obtain average inference times smaller than one millisecond when computing diagnostic queries using arithmetic circuits that model our real-world electrical power system.

Reference:

O. J. Mengshoel, A. Darwiche, K. Cascio, M. Chavira, S. Poll, and S. Uckun, “Diagnosing Faults in Electrical Power Systems of Spacecraft and Aircraft”, In Proc. of the Twentieth Innovative Applications of Artificial Intelligence, Conference (IAAI-08), Chicago, IL, 2008.

BibTex Reference:

@inproceedings{mengshoel08diagnosing, author = {Mengshoel, O. J. and Darwiche, A. and Cascio, K. and Chavira, M. and Poll, S. and Uckun, S.}, title = {Diagnosing Faults in Electrical Power Systems of Spacecraft and Aircraft}, booktitle = {Proceedings of the Twentieth Innovative Applications of Artificial Intelligence Conference (IAAI-08)}, pages = {1699--1705}, address = {Chicago, IL}, year = {2008} }