Development of a Swing-Tracking Sliding Mode Controller Design for Nonlinear Inverted Pendulum System via Bees-Slice Genetic Algorithm

Abstract

A new development of a swing-tracking control algorithm for nonlinear inverted pendulum system presents in this paper. Sliding mode control technique is used and guided by Lyapunov stability criterion and tuned by Bees-slice genetic algorithm (BSGA). The main purposes of the proposed nonlinear swing-tracking controller is to find the best force control action for the real inverted pendulum model in order to stabilize the pendulum in the inverted position precisely and quickly. The Bees-slice genetic algorithm (BSGA) is carried out as a stable and robust on-line auto-tune algorithm to find and tune the parameters for the sliding mode controller. Sigmoid function is used as signum function for sliding mode in order to eliminate the chattering effect of the fast switching surface by reducing the amplitude of the function output. MATLAB simulation results and LabVIEW experimental work are confirmed the performance of the proposed tuning swing-tracking control algorithm in terms of the robustness and effectiveness that is overcame the undesirable boundary disturbances, minimized the tracking angle error to zero value and obtained the smooth and best force control action for the pendulum cart, with fast and minimum number of fitness evaluation. INTRODUCTION owadays many research and dissertations consider the inverted pendulum system as a main topic of great interest due to their evolutionary application. Various fields such as rocket technology, missile launchers, air planes, ship, mobile robots and human walking use inverted pendulum [1 and 2]. Traditionally, the problems of inverted pendulum control have been addressed by stabilizing point of pendulum in the inverted position for inverted pendulum system, many control algorithm are proposed in the swing-tracking problems, such as back-stepping with sliding mode controller [3], hierarchical sliding mode controller [4], neural network with sliding mode controller [5], fuzzy with sliding mode controllers [6 and 7], adaptive-fuzzy with sliding mod controller [8], second-order sliding mode controller [9], PID with sliding mode controller [10], pole placement with sliding mode controller [11] , etc. This work, the cores of the motivation are taken from [4, 7, 12 and 13] because the problems in swing-tracking inverted pendulum are still there waiting to be processed in terms of generating best force control action for inverted pendulum cart, tracking the reference position with minimum swing-tracking error and overcoming undesirable disturbances. In this paper, the contributions can be described as follows:  The swing-tracking control law is developed through high analytical derive accuracy that it is based on sliding mode control, Lyapunov criterion stability and Bees-slice genetic algorithm (BSGA) in order to find the best force control action quickly.  The swing-tracking error of the inverted pendulum is minimized.