Advanced search
1 file | 1.36 MB Add to list

Towards a neural hierarchy of time scales for motor control

Author
Organization
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
Reservoir computing, Locomotion Control Hierarchy, Adaptive control, Central Pattern Generator, Feedback control

Downloads

  • sab2012.pdf
    • full text
    • |
    • open access
    • |
    • PDF
    • |
    • 1.36 MB

Citation

Please use this url to cite or link to this publication:

MLA
Waegeman, Tim, et al. “Towards a Neural Hierarchy of Time Scales for Motor Control.” LECTURE NOTES IN COMPUTER SCIENCE, edited by Tom Ziemke et al., vol. 7426, Springer Berlin, 2012, pp. 146–55, doi:10.1007/978-3-642-33093-3_15.
APA
Waegeman, T., wyffels, F., & Schrauwen, B. (2012). Towards a neural hierarchy of time scales for motor control. In T. Ziemke, C. Balkenius, & J. Hallam (Eds.), LECTURE NOTES IN COMPUTER SCIENCE (Vol. 7426, pp. 146–155). https://doi.org/10.1007/978-3-642-33093-3_15
Chicago author-date
Waegeman, Tim, Francis wyffels, and Benjamin Schrauwen. 2012. “Towards a Neural Hierarchy of Time Scales for Motor Control.” In LECTURE NOTES IN COMPUTER SCIENCE, edited by Tom Ziemke, Christian Balkenius, and John Hallam, 7426:146–55. Berlin, Germany: Springer Berlin. https://doi.org/10.1007/978-3-642-33093-3_15.
Chicago author-date (all authors)
Waegeman, Tim, Francis wyffels, and Benjamin Schrauwen. 2012. “Towards a Neural Hierarchy of Time Scales for Motor Control.” In LECTURE NOTES IN COMPUTER SCIENCE, ed by. Tom Ziemke, Christian Balkenius, and John Hallam, 7426:146–155. Berlin, Germany: Springer Berlin. doi:10.1007/978-3-642-33093-3_15.
Vancouver
1.
Waegeman T, wyffels F, Schrauwen B. Towards a neural hierarchy of time scales for motor control. In: Ziemke T, Balkenius C, Hallam J, editors. LECTURE NOTES IN COMPUTER SCIENCE. Berlin, Germany: Springer Berlin; 2012. p. 146–55.
IEEE
[1]
T. Waegeman, F. wyffels, and B. Schrauwen, “Towards a neural hierarchy of time scales for motor control,” in LECTURE NOTES IN COMPUTER SCIENCE, Odense, Denmark, 2012, vol. 7426, pp. 146–155.
@inproceedings{2976697,
  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.}},
  author       = {{Waegeman, Tim and wyffels, Francis and Schrauwen, Benjamin}},
  booktitle    = {{LECTURE NOTES IN COMPUTER SCIENCE}},
  editor       = {{Ziemke, Tom and Balkenius, Christian and Hallam, John}},
  isbn         = {{9783642330926}},
  issn         = {{0302-9743}},
  keywords     = {{Reservoir computing,Locomotion Control Hierarchy,Adaptive control,Central Pattern Generator,Feedback control}},
  language     = {{eng}},
  location     = {{Odense, Denmark}},
  pages        = {{146--155}},
  publisher    = {{Springer Berlin}},
  title        = {{Towards a neural hierarchy of time scales for motor control}},
  url          = {{http://doi.org/10.1007/978-3-642-33093-3_15}},
  volume       = {{7426}},
  year         = {{2012}},
}

Altmetric
View in Altmetric