Advertisement

Towards a Neural Hierarchy of Time Scales for Motor Control

  • Tim Waegeman
  • Francis Wyffels
  • Benjamin Schrauwen
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 7426)

Abstract

Animals show remarkable rich motion skills which are still far from realizable with robots. Inspired by the neural circuits which generate rhythmic motion patterns in the spinal cord of all vertebrates, one main research direction points towards the use of central pattern generators in robots. On of the key advantages of this, is that the dimensionality of the control problem is reduced. In this work we investigate this further by introducing a multi-timescale control hierarchy with at its core a hierarchy of recurrent neural networks. By means of some robot experiments, we demonstrate that this hierarchy can embed any rhythmic motor signal by imitation learning. Furthermore, the proposed hierarchy allows the tracking of several high level motion properties (e.g.: amplitude and offset), which are usually observed at a slower rate than the generated motion. Although these experiments are preliminary, the results are promising and have the potential to open the door for rich motor skills and advanced control.

Keywords

Locomotion Control Hierarchy Adaptive control Feedback control Central Pattern Generator Reservoir computing (RC) 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Kiebel, S., Daunizeau, J., Friston, K.: A hierarchy of time-scales and the brain. PLoS Computational Biology 4(11), e1000209 (2008)Google Scholar
  2. 2.
    Stein, P.: Neurons, networks, and motor behavior. The MIT Press (1999)Google Scholar
  3. 3.
    Ijspeert, A.: Central pattern generators for locomotion control in animals and robots: a review. Neural Networks 21(4), 642–653 (2008)CrossRefGoogle Scholar
  4. 4.
    Cohen, A., Wallen, P.: The neuronal correlate of locomotion in fish. Experimental Brain Research 41(1), 11–18 (1980)CrossRefGoogle Scholar
  5. 5.
    Rossignol, S.: Locomotion and its recovery after spinal injury. Current Opinion in Neurobiology 10(6), 708–716 (2000)CrossRefGoogle Scholar
  6. 6.
    Ijspeert, A., Crespi, A., Ryczko, D., Cabelguen, J.: From swimming to walking with a salamander robot driven by a spinal cord model. Science 315(5817), 1416–1420 (2007)CrossRefGoogle Scholar
  7. 7.
    Verstraeten, D., Schrauwen, B., D’Haene, M., Stroobandt, D.: An experimental unification of reservoir computing methods. Neural Networks 20, 391–403 (2007)zbMATHCrossRefGoogle Scholar
  8. 8.
    Jaeger, H.: The “echo state” approach to analysing and training recurrent neural networks. German National Research Center for Information Technology, Tech. Rep. GMD Report 148 (2001)Google Scholar
  9. 9.
    Maass, W., Natschläger, T., Markram, H.: Real-time computing without stable states: A new framework for neural computation based on perturbations. Neural Computation 14(11), 2531–2560 (2002)zbMATHCrossRefGoogle Scholar
  10. 10.
    Antonelo, E.A., Schrauwen, B., Stroobandt, D.: Event detection and localization for small mobile robots using reservoir computing. Neural Networks 21, 862–871 (2008)CrossRefGoogle Scholar
  11. 11.
    Jaeger, H., Haas, H.: Harnessing nonlinearity: predicting chaotic systems and saving energy in wireless telecommunication. Science 308, 78–80 (2004)CrossRefGoogle Scholar
  12. 12.
    Schrauwen, B., D’Haene, M., Verstraeten, D., Van Campenhout, J.: Compact hardware liquid state machines on FPGA for real-time speech recognition. Neural Networks 21, 511–523 (2008)CrossRefGoogle Scholar
  13. 13.
    Paquot, Y., Duport, F., Smerieri, A., Dambre, J., Schrauwen, B., Haelterman, M., Massar, S.: Optoelectronic reservoir computing. Scientific Reports 2, 1–6 (2012)CrossRefGoogle Scholar
  14. 14.
    Legenstein, R.A., Maass, W.: Edge of chaos and prediction of computational performance for neural microcircuit models. Neural Networks, 323–333 (2007)Google Scholar
  15. 15.
    Sussillo, D., Abbott, L.: Generating coherent patterns of activity from chaotic neural networks. Neuron 63(4), 544–557 (2009)CrossRefGoogle Scholar
  16. 16.
    Wyffels, F., Schrauwen, B.: Design of a central pattern generator using reservoir computing for learning human motion. In: Proceedings of the ECSIS Symposium on Advanced Technologies for Enhanced Quality of Life, pp. 118–122 (2009)Google Scholar
  17. 17.
    Waegeman, T., Wyffels, F., Schrauwen, B.: A discrete/rhythmic pattern generating RNN. In: Proceedings of the 20th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning, pp. 567–572. Ciaco - i6doc.com, Louvain-la-Neuve (2012)Google Scholar
  18. 18.
    Li, J., Jaeger, H.: Minimal energy control of an ESN pattern generator. Jacobs University, Tech. Rep. (2011)Google Scholar
  19. 19.
    Waegeman, T., Schrauwen, B.: Towards Learning Inverse Kinematics with a Neural Network Based Tracking Controller. In: Lu, B.-L., Zhang, L., Kwok, J. (eds.) ICONIP 2011, Part III. LNCS, vol. 7064, pp. 441–448. Springer, Heidelberg (2011)CrossRefGoogle Scholar
  20. 20.
    Waegeman, T., Wyffels, F., Schrauwen, B.: Feedback control by online learning an inverse model. IEEE Transactions on Neural Networks and Learning Systems (submitted)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2012

Authors and Affiliations

  • Tim Waegeman
    • 1
  • Francis Wyffels
    • 1
  • Benjamin Schrauwen
    • 1
  1. 1.Department of Electronics and Information SystemsGhent UniversityGhentBelgium

Personalised recommendations