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

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.

[1]  Mitsuo Kawato,et al.  Feedback-Error-Learning Neural Network for Supervised Motor Learning , 1990 .

[2]  Michael I. Jordan.  Supervised learning and systems with excess degrees of freedom , 1988 .

[3]  Michael A. Arbib,et al.  Perceptual Structures and Distributed Motor Control , 1981 .

[4]  Karl-Theodor Kalveram.  Power series and neural-net computing , 1993, Neurocomputing.

[5]  P. Matthews.  MUSCLE SPINDLES AND THEIR MOTOR CONTROL. , 1964, Physiological reviews.

[6]  P. Rudomín.  Presynaptic inhibition of muscle spindle and tendon organ afferents in the mammalian spinal cord , 1990, Trends in Neurosciences.

[7]  Mitsuo Kawato,et al.  Feedback-error-learning neural network for trajectory control of a robotic manipulator , 1988, Neural Networks.

[8]  P. Rudomin,et al.  Bulbospinal inhibition of PAD elicited by stimulation of afferent and motor axons in the isolated frog spinal cord and brainstem , 2005, Experimental Brain Research.

[9]  Kalveram Kt.  Pattern generating and reflex-like processes controlling aiming movements in the presence of inertia, damping and gravity : a theoretical note , 1991 .

[10]  C. G. Phillips,et al.  The Ferrier Lecture, 1968 - Motor apparatus of the baboon’s hand , 1969, Proceedings of the Royal Society of London. Series B. Biological Sciences.

[11]  P. Merton.  The silent period in a muscle of the human hand , 1951, The Journal of physiology.

[12]  Serge Rossignol,et al.  Phase-dependent modulation of dorsal root potentials evoked by peripheral nerve stimulation during fictive locomotion in the cat , 1990, Brain Research.

[13]  Mohsen Shahinpoor.  A robot engineering textbook , 1987 .

[14]  K. Kalveram,et al.  Pattern generating and reflex-like processes controlling aiming movements in the presence of inertia, damping and gravity , 1991, Biological Cybernetics.

[15]  E. Holst,et al.  Das Reafferenzprinzip , 2004, Naturwissenschaften.

[16]  L. Nashner.  Adapting reflexes controlling the human posture , 1976, Experimental Brain Research.

[17]  J. Kelso,et al.  Skilled actions: a task-dynamic approach. , 1987, Psychological review.

[18]  Karl-Theodor Kalveram,et al.  Controlling the dynamics of a two-joined arm by central patterning and reflex-like processing , 1991, Biological Cybernetics.