A non linear PID fuzzy approach to inverted pendulum controller with evolutionary parameter optimisation

Abstract

The balancing of an inverted pendulum by moving a cart along a horizontal track is a classic control problem. The direct approach to solve this problem is to use a control system which derives inputs from two observable parameters, which are error in the pole angle and deviation of cart position from the centre of the track. The proportional, integral and derivative aspects of these measures shall be combined to decide on the torque to be applied on the cart. The system is non-linear demanding a non-linear combination of the error measures for effective control. The approach proposed in this work is to achieve a non-linear control system using fuzzy modelling and the control system parameters are optimised using evolutionary computation. The results show the efficiency of the proposed method in achieving control over non-linear and unstable processes. The fuzzy modelling of the system offers better interpretability of the system making the approach extensible to a large variety of engineering problems.