[scottpoll71]

A Diagnostic Competition defined by NASA Ames Research Center and Palo Alto Research Center was hosted at the 20th International Workshop on Principles of Diagnosis (DX). It is proposed that this will continue with forthcoming DX events.

A Diagnostic Competition defined by NASA Ames Research Center and Palo Alto Research Center was hosted at the 20th International Workshop on Principles of Diagnosis (DX). It is proposed that this will continue with forthcoming DX events. The objectives of the competition are to accelerate research in theories, principles, and computational techniques for monitoring and diagnosis of complex systems, to encourage the development of software platforms that promise more rapid, accessible, and effective maturation of diagnostic technologies, and to provide a forum that can be utilized by algorithm developers to test and validate their technologies. The overall goal of this competition is to systematically evaluate different diagnostic technologies and to produce comparable performance assessments for different diagnostic methods.

Diagnostic algorithms competed in 3 tracks that included industrial and synthetic systems. Initially the participants were provided with system descriptions and a sample data set that included nominal and fault scenarios. The participants had to provide algorithms that communicated with the run-time architecture to receive scenario data and return diagnostic results. For the competition we ran all the algorithms on extended scenario data sets (different from sample set) to compute a set of pre-defined metrics and a final ranking based on weighted metric scores.

More details about and results from the competition can be found in the DX-09 proceedings, http://www.isy.liu.se/dx09/papers/proceedings-dx09.pdf (see proceedings page 371, pdf page 381) and also at the competition website http://www.dx-competition.org/. The website has more recent results from a “rematch” where diagnostic competition participants as well as all diagnostic algorithm developers from the community were invited to submit updates to algorithms or new algorithms for evaluation on competition scenarios in addition to some new scenarios created for the rematch.

[Author]

Posts