Intelligent decisions and dynamic coordination: properties of geometrical representation by generalized frames intrinsic to neural and robotic systems

Author

Pellionisz, A.J.

Author_Institution

Dept. of Physiol. & Biophys., New York Univ. Med. Center, NY, USA

fYear

1988

fDate

24-27 July 1988

Firstpage

603

Abstract

System neuroscience and robotics face comparable problems that can best be overcome if the two fields are unified by common concepts and formalisms. An approach to neurobotics is based on the axiom that the CNS functions by means of generalized natural coordinates that are intrinsic to the organism. Neural nets that embody internal geometrical representations of the external world and are composed of metric tensors of the sensory- and motor-spacetime domains, allow intelligent decisions on possible dynamic trajectories. By setting up a modifiable sequence of associations of trajectory-endpoints with commands, they also make possible trained and learned (skilled) performance.<>

Keywords

learning systems; neural nets; position control; robots; CNS; central nervous system; dynamic coordination; dynamic trajectories; generalized frames; generalized natural coordinates; geometrical representation; intelligent decisions; learned performance; metric tensors; modifiable sequence; motor-spacetime domains; neural nets; neurobotics; robotic systems; sensory-spacetime domain; skilled performance; system neuroscience; trained performance; trajectory-endpoints; Learning systems; Neural networks; Position control; Robots;

fLanguage

English

Publisher

ieee

Conference_Titel

Neural Networks, 1988., IEEE International Conference on

Conference_Location

San Diego, CA, USA

Type

conf

DOI

10.1109/ICNN.1988.23977

Filename

23977