Abstract
A fuzzy-Gaussian neural network (FGNN) controller is described by applying a Gaussian function as an activation function. A specialized learning architecture is used so that membership function can be tuned without using expert's manipulated data. As an example of the application, a tracking control problem for the speed and azimuth of a mobile robot driven by two independent wheels is solved by using the FGNN controller. To simplify the implementation of the FGNN controller for the two-input/two-output controlled system, a learning controller is utilized consisting of two FGNN's based on independent reasoning and a connection net with fixed weights. The effectiveness of the proposed method is illustrated by performing the simulation of a circular or square trajectory tracking control.
Original language | English |
---|---|
Pages (from-to) | 193-199 |
Number of pages | 7 |
Journal | IEEE Transactions on Control Systems Technology |
Volume | 4 |
Issue number | 2 |
DOIs | |
Publication status | Published - Dec 1 1996 |
Externally published | Yes |
ASJC Scopus subject areas
- Control and Systems Engineering
- Electrical and Electronic Engineering