Learning Multiple Models of Non-linear Dynamics for Control Under Varying Contexts

  • Georgios Petkos
  • Marc Toussaint
  • Sethu Vijayakumar
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4131)


For stationary systems, efficient techniques for adaptive motor control exist which learn the system’s inverse dynamics online and use this single model for control. However, in realistic domains the system dynamics often change depending on an external unobserved context, for instance the work load of the system or contact conditions with other objects. A solution to context-dependent control is to learn multiple inverse models for different contexts and to infer the current context by analyzing the experienced dynamics. Previous multiple model approaches have only been tested on linear systems. This paper presents an efficient multiple model approach for non-linear dynamics, which can bootstrap context separation from context-unlabeled data and realizes simultaneous online context estimation, control, and training of multiple inverse models. The approach formulates a consistent probabilistic model used to infer the unobserved context and uses Locally Weighted Projection Regression as an efficient online regressor which provides local confidence bounds estimates used for inference.


Hide Markov Model Inverse Model Multiple Model Unlabeled Data Proportional Derivative Control 
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.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Imamizu, H., Osu, R., Yoshioka, T., Flanagan, R., Nakano, E., Kawato, M.: Composition and decomposition of internal models in motor learning under altered kinematic and dynamic environments. The Journal of Neuroscience 19 (1999)Google Scholar
  2. 2.
    Wolpert, D.M., Haruno, M., Kawato, M.: Mosaic model for sensorimotor learning and control. Neural Computation 13, 2201–2220 (2001)MATHCrossRefGoogle Scholar
  3. 3.
    Wolpert, D.M., Kawato, M.: Multiple paired forward and inverse models for motor control. Neural Networks 11, 1317–1329 (1998)CrossRefGoogle Scholar
  4. 4.
    Narendra, K.S., Balakrishnan, J.: Adaptive control using multiple models. IEEE Transactions in automatic control 42, 171–187 (1997)MATHCrossRefMathSciNetGoogle Scholar
  5. 5.
    Narendra, K.S., Xiang, C.: Adaptive control of discrete-time systems using multiple models. IEEE Transactions in automatic control 45, 1669–1686 (2000)MATHCrossRefMathSciNetGoogle Scholar
  6. 6.
    D’Souza, A., Vijayakumar, S., Schaal, S.: Incremental online learning in high dimensions. Neural Computation 17, 2602–2634 (2005)CrossRefMathSciNetGoogle Scholar
  7. 7.
    Shibata, T., Conradt, J., Vijayakumar, S., D’Souza, A., Schaal, S.: Statistical learning for humanoid robots. Autonomous Robots 12, 55–69 (2002)MATHCrossRefGoogle Scholar
  8. 8.
    Verbeek, J.J., Vlassis, N., Kröse, B.J.A.: Efficient greedy learning of gaussian mixture models. Neural Computation 15(2), 469–485 (2003)MATHCrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Georgios Petkos
    • 1
  • Marc Toussaint
    • 1
  • Sethu Vijayakumar
    • 1
  1. 1.Institute of Perception, Action and Behaviour, School of InformaticsUniversity of Edinburgh

Personalised recommendations