Analysis of Control of Inverted Pendulum using Adaptive Neuro Fuzzy System

Abstract

In this thesis modeling and simulation study of basically two control strategies of an inverted pendulum system are presented. The goal here is to determine which control strategy delivers better performance with respect to pendulum’s angle and cart’s position. The inverted pendulum represents a challenging control problem, which continually moves toward an uncontrolled state. Certain modern techniques are now available because of the development of artificial intelligence and soft computing methods. These methods make the system user friendly and precise control becomes an additional feature. Fuzzy control based learning makes the system more reliable and taking the advantage of precise control at different levels of existence as fuzzy is applicable to non value zero to full value as one, rest of the interim values are also considered. ANFIS is one of an example of fused neuro fuzzy system in which the fuzzy system is incorporated in such a framework which is adaptive in nature. The fuzzy controllers are used to sort out the problem of learning and when the ANFIS is used the adaptability of the system get improves which is based on that learning feature for non-linear system of inverted pendulum model. Learning, training and consideration of the interim levels of the fractional values of variables makes the system sophisticated. Facility of modelling makes the system workable. Modelling is done using Matlab and Simulation study has been done in Simulink in the inverted pendulum case study undertaken. It is found that ANFIS produces better modelling strategy as compared to fuzzy logical controllers (FLC).