Searching a Scalable Approach to Cerebellar Based Control

Authors: Peters J.¹; van der Smagt P.²

Source: Applied Intelligence, Volume 17, Number 1, July 2002 , pp. 11-33(23)

Publisher: Springer

Buy & download fulltext article:

Price: $47.00 plus tax (Refund Policy)

Abstract:

Decades of research into the structure and function of the cerebellum have led to a clear understanding of many of its cells, as well as how learning might take place. Furthermore, there are many theories on what signals the cerebellum operates on, and how it works in concert with other parts of the nervous system. Nevertheless, the application of computational cerebellar models to the control of robot dynamics remains in its infant state. To date, few applications have been realized.

The currently emerging family of light-weight robots (Hirzinger, in Proc. Second ecpd Int. Conference on Advanced Robotics, Intelligent Automation and Active Systems, 1996) poses a new challenge to robot control: due to their complex dynamics traditional methods, depending on a full analysis of the dynamics of the system, are no longer applicable since the joints influence each other dynamics during movement. Can artificial cerebellar models compete here?

Keywords: robot dynamics; robot arm control; computational cerebellar models; neural networks

Language: English

Document Type: Regular paper

Affiliations: 1: Computational Learning and Motor Control Lab, University of Southern California, 3641 Watt Way, Los Angeles, CA 90. jrpeters@usc.edu 2: Institute of Robotics and Mechatronics, German Aerospace Center, DLR Oberpfaffenhofen, P.O. Box 1116, 82230 Wessling, Germany. smagt@dlr.de

Publication date: 2002-07-01