Particle Swam Optimization for Stabilizing Controller of a Self-erecting Linear Inverted Pendulum

This paper presents an advanced application of particle swarm optimization, PSO to find state feedback controller gains for stabilizing controller in a linear inverted pendulum. This plant is used as an application example of the proposed method. In conventional method of state feedback control design such as pole placement and linear quadratic regulator method, controller designers often face troublesome exercise of tuning several parameters. Particularly, one has to face trial-and- error approach to select suitable Q and R matrices to design a state feedback control using linear quadratic regulator method. To overcome this problem, an intelligent approach employing PSO- based constrained optimization is proposed. The objective of the optimization is to minimize error function, while closed loop poles region is incorporated as an optimization constraint whose parameter is selected based on the desired control performance. In this study, Clerc's PSO is adopted together with dynamic objective constraint handling where efficient optimization run is shown in the simulation results.