Please select the desired project time frame:

• July 2026
• January 2026
• July 2025
• January 2025
• July 2024
• January 2024
• July 2023
• January 2023
• July 2022
• January 2022
• July 2021
• January 2021
• July 2020
• January 2020
• July 2019
• January 2019
• July 2018
• January 2018
• July 2017
• January 2017
• July 2016
• January 2016
• July 2015
• January 2015
• July 2014
• January 2014
• July 2013
• January 2013
• July 2012
• January 2012
• July 2011
• January 2011
• July 2010
• January 2010
• July 2009
• January 2009
• July 2008
• January 2008
• July 2007
• January 2007
• July 2006
• January 2006
• July 2005
• January 2005
• July 2004
• January 2004
• July 2003
• January 2003
• July 2002
• January 2002
• July 2001
• January 2001

Start of funding 01.01.2009

Planning, Diagnosis and Learning in Reconfigurable Production Systems

---

Manufacturing is increasingly evolving towards flexible production processes, where factories reconfigure and optimize themselves for individual products. Controlling such systems poses new challenges, because production plans must be modified on-the-fly, and the overall uncertainty of plan execution increases. In our cooperation, we seek to combine expertise from TUM and PARC in artificial intelligence, model-based reasoning and diagnosis technology, in order to adapt and evaluate methods for self-planning, self-diagnosis and self-learning in the context of flexible production systems. The intended output of this collaboration will be a prototype of a self-reconfigurable machine shop implemented in TUM’s cognitive factory test-bed, which uses these methods to automatically adjust to changes and recover from anomalous situations during execution.

Final report:
The goal of this project was to develop novel algorithmic methods for self-planning, self-diagnosis and learning in the context of flexible production systems. Our main application test-bed was the cognitive machine shop that is part of TUM’s CoTeSys cognitive factory. The initial funding of the project was used for an extended research stay of the junior scientist Paul Maier in 2009, and a visit and invited talk of Martin Sachenbacher at PARC in 2010. In exchange, PARC junior scientist Lukas Kuhn visited TUM in 2009 and 2010. As a result of the project, a model-based capability has been developed for integrated diagnosis and prognosis that can automatically detect and assess possible anomalies early on and adapt production plans if necessary. The results have been documented in several international publiations [1,2,3] and form a major part of the Paul Maier’s dissertation. The project subsequently led to a Dagstuhl seminar on Runtime Verification, Diagnosis, Planning and Control for Autonomous Systems [4], held in in 2010 with the participation of Johan de Kleer.

References

[1] Paul Maier, Martin Sachenbacher, Thomas Ru¨hr, and Lukas Kuhn: Integrating Model-based Diagnosis and Prognosis in Autonomous Production. IEEE International Conference on Prognostics and Health Management (PHM), San Diego, USA, 2009.

[2] Paul Maier, Dominik Jain, Stefan Waldherr, and Martin Sachenbacher: Plan Assessment for Autonomous Manufacturing as Bayesian Inference, German Conference on Artificial Intelligence (KI), Karlsruhe, Germany, 2010.

[3] Paul Maier, Martin Sachenbacher, Thomas Ru¨hr, and Lukas Kuhn: Automated Plan Assessment in Cognitive Manufacturing. Advanced Engineering Informatics 24(4):308-319, 2010.

[4] Klaus Havelund, Martin Leucker, Martin Sachenbacher, Oleg Sokolsky, and Brian C. Williams (eds.): Runtime Verification, Diagnosis, Planning and Control for Autonomous Systems. Dagstuhl Seminar Proceedings 10451, Schloss Dagstuhl - Leibniz-Zentrum fu¨r Informatik, Dagstuhl, Germany, 2011.