A PID based ANFIS control of inverted pendulum climbing on inclined plane

Abstract

Inverted pendulum on cart is a very popular non-linear, nonholonomic and multi-variable system which exhibits highly complex dynamics which attracts attention of many researchers. Control of inverted pendulum is a difficult task due to large number of uncertainties associated with the system. This paper presents an attempt to control a variant of inverted pendulum which is climbing up an inclined plane. An angle of inclination induced into the system makes control of pendulum system more difficult and challenging. In this study a Proportional-integral-derivative (PID) controller is designed for offline mode control of Inverted pendulum on Inclined plane (IPIP). The objective is to achieve and maintain the desired position and upright orientation of proposed system within specified time. A Matlab-Simulink model of the proposed system has been constructed from derived mathematical equations. The data sets generated from PID controller were further used to train an Adaptive neuro fuzzy inference system (ANFIS) controller. The simulation results showed the excellent learning ability of ANFIS controller. The performance parameters considered for comparison of controllers were settling time, maximum overshoot and steady state error.