Learning in a High Dimensional Space: Fast Omnidirectional Quadrupedal Locomotion

  • Matthias Hebbel
  • Walter Nistico
  • Denis Fisseler
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4434)


This paper presents an efficient way to learn fast omnidirectional quadrupedal walking gaits. We show that the common approaches to control the legs can be further improved by allowing more degrees of freedom in the trajectory generation for the legs. To achieve good omnidirectional movements, we suggest to use different parameters for different walk requests and interpolate between them. The approach has been implemented for the Sony Aibo and used by the GermanTeam in the Four-Legged-League in 2005. A standard learning strategy has been adopted, so that the optimization process of a parameter set can be done within one hour, without human intervention. The resulting walk achieved remarkable speeds, both in pure forward walking and in omnidirectional movements.


High Dimensional Space Normal Walk Quadruped Robot Fast Walk Legged Robot 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


  1. 1.
    Hengst, B., Ibbotson, D., Pham, S.B., Sammut, C.: Omnidirectional locomotion for quadruped robots. In: Birk, A., Coradeschi, S., Tadokoro, S. (eds.) RoboCup 2001. LNCS (LNAI), vol. 2377, pp. 368–373. Springer, Heidelberg (2002)CrossRefGoogle Scholar
  2. 2.
    Düffert, U., Hoffmann, J.: Reliable and precise gait modeling for a quadruped robot. In: Bredenfeld, A., Jacoff, A., Noda, I., Takahashi, Y. (eds.) RoboCup 2005. LNCS (LNAI), vol. 4020, Springer, Heidelberg (2006)CrossRefGoogle Scholar
  3. 3.
    Kohl, N., Stone, P.: Policy gradient reinforcement learning for fast quadrupedal locomotion. In: ICRA. Proceedings of the IEEE International Conference on Robotics and Automation, IEEE Computer Society Press, Los Alamitos (2004)Google Scholar
  4. 4.
    Röfer, T., Laue, T., Burkhard, H., Hoffmann, J., Jüngel, M., Göhring, D., Lötzsch, M., Spranger, M., Altmeyer, B., Goetzke, V., von Stryk, O., Brunn, R., Dassler, M., Kunz, M., Risler, M., Stelzer, M., Thomas, D., Uhrig, S., Schwiegelshohn, U., Dahm, I., Hebbel, M., Nistico, W., Schumann, C., Wachter, M.: German Team Report 2004. Technical report, HU Berlin, TU Bremen, TU Darmstadt and University of Dortmund (2004)Google Scholar
  5. 5.
    Stone, P., Dresner, K., Fidelman, P., Jong, N.K., Kohl, N., Kuhlmann, G., Sridharan, M., Stronger, D.: The UT Austin Villa 2004 RoboCup four-legged team: Coming of age. Technical Report UT-AI-TR-04-313, The University of Texas at Austin, Department of Computer Sciences, AI Laboratory (2004)Google Scholar
  6. 6.
    Cohen, D., Ooi, Y.H., Vernaza, P., Lee, D.D.: Robocup 2004 Legged Soccer Team. Technical report, University of Pennsylvania (2004)Google Scholar
  7. 7.
    Kohl, N., Stone, P.: Machine learning for fast quadrupedal locomotion. In: The Nineteenth National Conference on Artificial Intelligence, pp. 611–616 (2004)Google Scholar
  8. 8.
    Röfer, T.: Evolutionary gait-optimization using a fitness function based on proprioception. In: Nardi, D., Riedmiller, M., Sammut, C., Santos-Victor, J. (eds.) RoboCup 2004. LNCS (LNAI), vol. 3276, Springer, Heidelberg (2005)Google Scholar
  9. 9.
    Craig, J.J.: 6, 7. In: Introduction to Robotics, Mechanics and Control, Addison Wesley, Reading (1986)Google Scholar
  10. 10.
    Dahm, I., Fisseler, D., Hebbel, M., Nistico, W.: Learning fast walking patterns with reliable odometry information for four-legged robots. In: ISMCR 2005 (2005)Google Scholar
  11. 11.
    Beyer, H.G., Schwefel, H.P.: Evolution strategies – A comprehensive introduction. Natural Computing 1(1), 3–52 (2002)zbMATHCrossRefMathSciNetGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2007

Authors and Affiliations

  • Matthias Hebbel
    • 1
  • Walter Nistico
    • 1
  • Denis Fisseler
    • 1
  1. 1.Robotics Research Institute, Information Technology Section, Universität Dortmund, Otto-Hahn-Str.8, 44227 Dortmund 

Personalised recommendations