InfoMax Bayesian learning of the Furuta pendulum

László A. Jeni, György Flórea, András Lorincz

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

We have studied the InfoMax (D-optimality) learning for the two-link Furuta pendulum. We compared InfoMax and random learning methods. The InfoMax learning method won by a large margin, it visited a larger domain and provided better approximation during the same time interval. The advantages and the limitations of the InfoMax solution are treated.

Original languageEnglish
Pages (from-to)637-649
Number of pages13
JournalActa Cybernetica
Volume18
Issue number4
Publication statusPublished - Jan 1 2008

    Fingerprint

Keywords

  • D-optimality
  • Furuta pendulum
  • Infomax control
  • Online Bayesian learning

ASJC Scopus subject areas

  • Software
  • Computer Science (miscellaneous)
  • Computer Vision and Pattern Recognition
  • Management Science and Operations Research
  • Information Systems and Management
  • Electrical and Electronic Engineering

Cite this

Jeni, L. A., Flórea, G., & Lorincz, A. (2008). InfoMax Bayesian learning of the Furuta pendulum. Acta Cybernetica, 18(4), 637-649.