Stage-oe-small.jpg

Phdthesis74

Aus Aifbportal
Wechseln zu:Navigation, Suche
Controlled Self-Organisation Using Learning Classifier Systems




Datum: 30. Juli 2009
PhD thesis at the Universität Karlsruhe (TH), Fakultät für Wirtschaftswissenschaften
Referent(en): Prof. Dr. H. Schmeck / Prof. Dr. K.-H. Waldmann
BibTeX


Kurzfassung
The complexity of technical systems continuously increases, breakdowns and fatal errors occur quite often. Therefore, the mission of organic computing is to tame these challenges by providing appropriate degrees of freedom for self-organised behaviour. Technical systems should adapt to changing requirements of their execution environment, in particular with respect to human needs. To achieve these ambitious goals, adequate methods and techniques have to be developed. The proposed generic observer/controller architecture constitutes one way to achieve controlled self-organisation in technical systems.

To improve the design of organic computing systems, the observer/controller architecture is applied to multi-agent scenarios from the predator/prey domain, which serve as testbeds for evaluation. Thereby, the aspect of on-line learning using learning classifier systems is specially addressed.

ISBN: 978-3-86644-431-7
Weitere Informationen unter: Link

Projekt

OCCSQE



Forschungsgruppe

Effiziente Algorithmen


Forschungsgebiet

Organic Computing, Maschinelles Lernen, Agentensysteme