Sudarat Khwanon, Thanudchai Kulworawanichpong and Saravut Sujitjorn

pp. 101 - 121

Abstract

This article presents a new neuro-fuzzy controller to stabilize an inverted pendulum system. The
proposed controller consists of the Single Input Rule Modules (SIRMs), the artificial neural network (ANN)
and the dynamic importance degrees (DIDs). It simultaneously controls both the angle of the pendulum and
the position of the cart. The learning of the ANN results in the DIDs. The proposed controller has a simple
structure that can decrease the number of fuzzy rules. The simulation results show that the proposed neurofuzzy controller has an ability to stabilize a wide range of the inverted pendulum system within a short period
of time. Moreover, the comparisons of the simulation results between the proposed neuro-fuzzy controller
and the SIRMs fuzzy controller are revealed in this article.

Keywords

• Single Input Rule Module (SIRM)
• Dynamic Importance Degree (DID)
• Artificial Neural Network (ANN)
• Fuzzy Logic