Dynamic gait pattern generation with reinforcement learning

Mustafa Suphi Erden, Kemal Leblebicioglu

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

This paper presents the gait pattern generation work performed for the sixlegged robot EA308 developed in our laboratory. The aim is to achieve a dynamically developing gait pattern generation structure using reinforcement learning. For the six legged robot a simplified simulative model is constructed. The algorithm constructs a radial basis function neural network (RBFNN) to command proper leg configurations to the simulative robot. The weights of the RBFNN are learned using reinforcement learning. The developed structure succeeded in learning gait patterns compatible with different speeds of the robot.

Original languageEnglish
Title of host publicationIFAC Proceedings Volumes (IFAC-PapersOnline)
Pages151-156
Number of pages6
Volume16
DOIs
Publication statusPublished - Dec 2005
Event16th Triennial World Congress of International Federation of Automatic Control - Prague, United Kingdom
Duration: 3 Jul 20058 Jul 2005

Conference

Conference16th Triennial World Congress of International Federation of Automatic Control
Abbreviated titleIFAC 2005
CountryUnited Kingdom
CityPrague
Period3/07/058/07/05

Keywords

  • Gait pattern
  • Radial basis function neural network
  • Reinforcement learning
  • Six-legged robot
  • Walking

Fingerprint Dive into the research topics of 'Dynamic gait pattern generation with reinforcement learning'. Together they form a unique fingerprint.

  • Cite this

    Erden, M. S., & Leblebicioglu, K. (2005). Dynamic gait pattern generation with reinforcement learning. In IFAC Proceedings Volumes (IFAC-PapersOnline) (Vol. 16, pp. 151-156) https://doi.org/10.3182/20050703-6-CZ-1902.01295