Seminar

Seminar Series No. 10: Fuzzy Logic and Neural Network for Industrial Robot Control

Assoc. Prof. Dr. Muhammad Suzuri Hitam


Date: 2008-07-31
Time: 15:00 to 16:00
Venue: Plenary Theatre


Abstract

There are many reasons why the control of a robot arm is difficult. The principal causes of this difficulty is the continuously varying arm configuration, uncertain effects of load handling on the stability of the arm, the impact of gravitational, inertia, coriolis and centrifugal forces on the dynamic coupling between joints, and finally structural resonance. In a robotic application, a precise mathematical model is very difficult to obtain if not impossible. The dynamic behaviour of a robot will change dramatically with the variations in the configuration of the links and the size of the payload. The severity of the control problem does increase significantly when the robot arm is operated at higher speeds. In this seminar, the application of fuzzy logic, the neural network and a combination of these two methods as an alternative to the existing methods for the control of an actual industrial robot will be presented. Since this project is based on a real process, the relevant issues and difficulties arising in the real-time implementation of the algorithms will be discussed. Suggestions for the practical solution of the problems that arose in this project are made which will be of interest to both the researcher and industrial practitioner alike.

About the Speaker