A hybrid learning architecture based on neural networks for adaptive control of a walking machine

Author

Ilg, Winfried ; Mühlfriedel, Thomas ; Berns, Karsten

Author_Institution

Gruppe Interaktive Planungstechnik, Forschungszentrum Inf., Karlsruhe, Germany

Volume

3

fYear

1997

fDate

20-25 Apr 1997

Firstpage

2626

Abstract

Online learning of complex control behaviour of autonomous mobile robots is one of the current research topics. In this article a hybrid learning architecture based on self-organizing neural networks for online adaptivity is presented. The hybrid concept integrates different learning methods and task-oriented representations as well as available domain knowledge. The proposed concept is used for reinforcement learning of control strategies on different control levels on a walking machine

Keywords

adaptive control; feedforward neural nets; learning (artificial intelligence); legged locomotion; mobile robots; motion control; neurocontrollers; recurrent neural nets; robot kinematics; self-adjusting systems; adaptive control; autonomous mobile robots; hybrid learning architecture; kinematics; online learning; radial basis function networks; recurrent neural nets; reinforcement learning; self-organizing neural networks; walking machine; Adaptive control; Control systems; Inspection; Learning systems; Leg; Legged locomotion; Machine learning; Mobile robots; Neural networks; Supervised learning;

fLanguage

English

Publisher

ieee

Conference_Titel

Robotics and Automation, 1997. Proceedings., 1997 IEEE International Conference on

Conference_Location

Albuquerque, NM

Print_ISBN

0-7803-3612-7

Type

conf

DOI

10.1109/ROBOT.1997.619357

Filename

619357