Sidia: Extending prediction based diagnosis to dynamic models

  • Thomas Guckenbiehl
  • Gisela Schäfer-Richter
Model-Based Diagnosis
Part of the Lecture Notes in Computer Science book series (LNCS, volume 462)


The paper summarizes results of a study about how the ‘general diagnostic engine’ (GDE) may be used to diagnose complex, dynamically modelled systems. To deal with complexity we extended GDE to hierarchical models and integrated filters for conflicts and candidates. Diagnosis at different levels in the hierarchy rendered our application tractable. Conflict and candidate filtering assumptions, retractable when necessary, were a good means for efficiently pruning the candidate space in typical situations, without loosing GDE's principal ability to diagnose unexpected faults. Concerning the dynamics in our application, a special extension of episode propagation was developed, and evaluation of probes had to be adapted. All of these extensions to GDE proved easily integratable without touching the basic mechanisms of prediction-based diagnosis. In particular, GDE's clear separation between diagnosis and behavior prediction allowed the straightforward integration of the new predictive engine. However, our work also showed that the application of GDE to complex and dynamic models requires further elaboration of some of its basic features. In particular, GDE's information theoretic probe selection procedure and the predictive engine should be supported by stronger heuristic knowledge.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [Decker 88]
    Rainer Decker: ”Modeling the temporal behavior of technical systems”, Proc. German Workshop on Artificial Intelligence GWAI, 1988, pp. 41–50.Google Scholar
  2. [Davis 84]
    Randy Davis: ”Diagnostic Reasoning based on Structure and Behavior”, Artificial Intelligence (24), 1984, pp. 347–410.Google Scholar
  3. [de Kleer 86]
    Johan de Kleer: ”An Assumption Based Truth Maintenance System”, Artificial Intelligence (28), 1986, pp. 127–162.Google Scholar
  4. [de Kleer / Williams 87]
    Johan de Kleer and Brian Williams: ”Diagnosing Multiple Faults”, Artificial Intelligence (32), 1987, pp. 97–130.Google Scholar
  5. [de Kleer / Willians 89]
    Johan de Kleer and Brian Williams: ”Diagnosis with Behavioral Modes”, Proc. IJCAI-89, 1989, pp. 1324–1330.Google Scholar
  6. [Genesereth 84]
    Michael Genesereth: ”The Use of Design Descriptions in Automated Diagnosis”, Artificial Intelligence (24), 1984, pp. 411–436.Google Scholar
  7. [Guckenbiehl 90]
    Thomas Guckenbiehl: The Extended Episode Propagator, FhG-IITB, internal report, august 1990.Google Scholar
  8. [Hamscher 88]
    Walter Hamscher: Model-Based Troubleshooting of Digital Systems, technical report 1074, MIT AI-lab, 1988.Google Scholar
  9. [Hamscher / Davis 84]
    Walter Hamscher and Randy Davis: ”Diagnosing Circuits with state: an inherently underconstrained problem”, Proc. AAAI-84, pp. 142–147.Google Scholar
  10. [Hamscher / Davis 87]
    Walter Hamscher and Randy Davis: Issues in Model Based Troubleshooting, MIT-AI-Memo 893, March 1987.Google Scholar
  11. [Motoda et al. 84]
    H.Motoda, N. Yamada, K. Yoshida: ”A knowledge based system for plant diagnosis”, Proc. International Conference on Fifth Generation Computer Systems, 1984, pp. 582–588.Google Scholar
  12. [Shoham 87]
    Yoav Shoham: Reasoning about Change: Time and Causation from the Standpoint of AI, Ph. D. thesis, Yale University, 1987.Google Scholar
  13. [Struss / Dressler 89]
    Peter Struss and Oskar Dressler: ”Physical Negation: The integration of Fault Models into the General Diagnostic Engine”, Proc. IJCAI-89, pp. 1318–1323.Google Scholar
  14. [Williams 86]
    Brian Wiliams: ”Doing Time: Putting Qualitative Reasoning on firmer ground”, Proc. AAAI-86, pp. 105–112.Google Scholar
  15. [Zeigler 84]
    Bernard Zeigler: Theory of Modelling and Simulation, John Wiley & sons, 1976.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1990

Authors and Affiliations

  • Thomas Guckenbiehl
    • 1
  • Gisela Schäfer-Richter
    • 1
  1. 1.Fraunhofer-Institute of Information and Data Processing (FhG-IITB)Karlsruhe

Personalised recommendations