Advertisement

Standing up with Motor Primitives

  • V. Hamburger
  • K. Berns
  • F. Iida
  • R. Pfeifer
Conference paper

Abstract

As observed in nature, complex locomotion can be generated based on an adequate combination of motor primitives. In this context, the paper focused on experiments which result in the development of a quality criterion for the design and analysis of motor primitives. First, the impact of different vocabularies on behavioural diversity, robustness of prelearned behaviours and learning process is elaborated. The experiments are performed with the quadruped robot MiniDog6M for which a running and standing up behaviour is implemented. Further, a reinforcement learning approach based on Q-learning is introduced which is used to select an adequate sequence of motor primitives.

Keywords

Motor primitive morphology behavioural diversity reinforcement learning quadruped locomotion 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Verena Hamburger. Locomotion of a quadruped robot based on motor primitives. Diploma thesis, Robotic System Group, TU Kaiserslautern, April 2005Google Scholar
  2. 2.
    Rolf Pfeifer, Christian Scheier. Understanding Intelligence. MIT Press, 1999.Google Scholar
  3. 3.
    E. Bizzi, N. Acornero, W. Chapple, N. Hogan. Posture control and trajectory formation during arm movement. 4:2738–2744, 1984.Google Scholar
  4. 4.
    Feldman. Superposition of motor programs, I. Rhythmic forearm movement in man. 5:81–90, 1980.Google Scholar
  5. 5.
    Simon F. Giszter, Fernando A. Muss-Ivaldi, Emilio Bizzi. Convergent Force Fields Organized in the Frog’s Spinal Cord. Journal of Neuroscience, Februar 1993.Google Scholar
  6. 6.
    F.A. Mussa-Ivaldi, S.F. Giszter, E. Bizzi. Linear combinations of primitives in vertebrate motor control. 1994.Google Scholar
  7. 7.
    Ansgar Büschges. Sensory Control and Organization of Neural Networks Mediating Coordination of Multisegmental Organs Organs for Locomotion. Journal of Neurophysiology, Februar 2005.Google Scholar
  8. 8.
    Gerhard Neumann. The Reinforcement Learning Toolbox, Reinforcement Learning for Optimal Control Tasks. Diploma thesis, Institute of Computer Science, TU Graz, May 2005.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • V. Hamburger
    • 1
  • K. Berns
    • 1
  • F. Iida
    • 2
  • R. Pfeifer
    • 2
  1. 1.AG RobotersystemeTU KaiserslauternGermany
  2. 2.Artificial Intelligence LaboratoryUniversity of ZürichSwitzerland

Personalised recommendations