The 21st SICE Kyushu Chapter Annual Conference Abstract [103C]

Last update: Fri Mar 28 21:23:56 2003

103C5
Internal NN Model Control for Nonholonomic system
AUTHORS
Yuki¡¤ Koga, Keigo¡¤Watanabe, Kiyotaka¡¤Izumi, Kazuo¡¤Kiguti (Saga University)
ABSTRACT
In this paper, we describe a method of control of nonholonomic systems using feedback error learning method. We clarify the similarity and difference between the neurointerface method recently developed by Widrow and the feedback error learning method, through the conventional internal model control (IMC). In particular, a disturbance rejection approach based on the feedback error learning method is proposed for a nonholonomic mobile robot when the change of the initial configuration and payload is regarded as a disturbance.