Biological Cybernetics

, Volume 68, Issue 2, pp 183–191

A neural network model rapidly learning gains and gating of reflexes necessary to adapt to an arm's dynamics

  • K. Th. Kalveram
Article

Abstract

Effects of dynamic coupling, gravity, inertia and the mechanical impedances of the segments of a multi-jointed arm are shown to be neutralizable through a reflex-like operating three layer static feedforward network. The network requires the proprioceptively mediated actual state variables (here angular velocity and position) of each arm segment. Added neural integrators (and/or differentiators) can make the network exhibit dynamic properties. Then, actual feedback is not necessary and the network can operate in a pure feedforward fashion. Feedforward of an additional load can easily be implemented into the network using “descendent gating”, and a negative feedback control loop added to the feedforward control reduces errors due to external noise. A training, which combines a least squared error based simultaneous learning rule (LSQ-rule) with a “self-imitation algorithm” based on direct inverse modeling, enables the network to acquire the whole inverse dynamics, limb parameters included, during one short training movement. The considerations presented also hold for multi-jointed manipulators.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Arbib MA (1981) Perceptual structures and distributed motor control. In: Brooks VB (eds) Motor control. Part 1. In: Handbook of physiology, Sect. 1: The nervous system, vol. II. American Physiological Society, Bethesda, pp 1449–1480Google Scholar
  2. Gonzales H, Jiminez I, Rudomin P (1992) Reticulospinal inhibition of PAD elicited by stimulation of afferent and motor axons in the isolated frog neuraxis. Exp Brain Res 86:106–116Google Scholar
  3. Gossard JP, Rossignol S (1990) Phase-dependent modulation of dorsal root potentials evoked by peripheral nerve stimulation during fictive locomotion in the cat. Brain Res 537:1–13Google Scholar
  4. Holst E von, Mittelstaedt H (1950) Das Reafferenzprinzip. Naturwissenschaften 37:464–476Google Scholar
  5. Jordan MI (1988) Supervised learning and systems with excess degrees of freedom. COINS Tech Rep 88-27:1–41Google Scholar
  6. Kalveram K Th (1981) Erwerb sensumotorischer Koordinationen unter störenden Umwelteinflüssen: Ein Beitrag zum Problem des Erlernens von Werkzeuggebrauch (Engl. title: Acquiring sensori-motor coordinations under disturbed feedback. A contribution to the problem of learning to handle a tool). In: Tent L (Hr) Erkennen, Wollen, Handeln. Festschrift für Heinrich Düker. Hogrefe, Göttingen, pp 336–348Google Scholar
  7. Kalveram K Th (1990) A neural network acquiring the inverse of a tool or limb tranformation by “self-imitation learning”. Poster at the ICNC-10th Cybernetic Congress of the DGK, Duesseldorf. (Reprint available on request from the author)Google Scholar
  8. Kalveram K Th (1991a) Pattern generating and reflex-like processes controlling aiming movements in the presence of inertia, damping and gravity. A theoretical note. Biol Cybern 64:413–419Google Scholar
  9. Kalveram K Th (1991b) Controlling the dynamics of a two-jointed arm by central patterning and reflex-like processing. Biol Cybern 65:65–71Google Scholar
  10. Kalveram K Th (1992) Power series and neural-net computing. Neurocomputing (in press)Google Scholar
  11. Kawato M (1991) Feedback-error-learning neural network for supervised motor learning. In: Eckmiller R (eds) Advanced neural computers. Elsevier, North Holland, pp 365–372Google Scholar
  12. Matthews PCB (1964) Muscle spindles and their motor control. Physiol Rev 44:219–288Google Scholar
  13. Merton PA (1951) The silent period in a muscle of the human hand. J Physiol London 114:183–198Google Scholar
  14. Miyamoto H, Kawato M, Setoyama T, Suzuki R (1988) Feedback-error-learning neural network for trajectory control of a robotic manipulator. Neural Networks 1:251–265Google Scholar
  15. Nashner LM (1976) Adapting reflexes controlling the human posture. Exp Brain Res 26:59–72Google Scholar
  16. Phillips CG (1969) Motor apparatus of the baboon's hand. Proc R Soc London Ser B 173:141–174Google Scholar
  17. Rudomin P (1990) Presynaptic inhibition of muscle spindle and tendon organ afferents in the mamalian spinal cord. Trends Neurosci 13:499–505Google Scholar
  18. Saltzman E, Kelso JAS (1987) Skilled actions: A task dynamic approach. Psychol Rev 94:84–106Google Scholar
  19. Shahinpoor M (1987) A robot engineering textbook. Harper & Row, New YorkGoogle Scholar

Copyright information

© Springer-Verlag 1992

Authors and Affiliations

  • K. Th. Kalveram
    • 1
  1. 1.Abteilung für Kybernetische Psychologie, Heinrich-Heine-UniversitätDüsseldorfFederal Republic of Germany

Personalised recommendations