Evolution and analysis of mixed mode neural networks for walking: mixed pattern generators

Author

Gallagher, John C.

Author_Institution

Dept. of Comput. Sci. & Eng., Wright State Univ., Dayton, OH, USA

Volume

1

fYear

2001

fDate

2001

Firstpage

397

Abstract

This paper summarizes the results of the dynamical systems analysis of nearly four hundred continuous-time recurrent neural network (CTRNN) single-leg locomotion controllers evolved under conditions where sensory information was unreliable and in which the body the controller was embedded in could change its physical properties. The general principles underlying the operation of all the resulting mixed pattern generators (MPGs) are discussed. Several MPG operational features are explained and verified. Finally, discussion is made of future extensions of this research

Keywords

genetic algorithms; legged locomotion; motion control; recurrent neural nets; continuous-time recurrent neural network; dynamical systems analysis; genetic algorithm; mixed mode neural networks; mixed pattern generators; sensory information; single legged locomotion controllers; walking; Computer science; Control systems; Information analysis; Legged locomotion; Neural networks; Neurons; Optimal control; Pattern analysis; Recurrent neural networks; Weight control;

fLanguage

English

Publisher

ieee

Conference_Titel

Evolutionary Computation, 2001. Proceedings of the 2001 Congress on

Conference_Location

Seoul

Print_ISBN

0-7803-6657-3

Type

conf

DOI

10.1109/CEC.2001.934418

Filename

934418