Analysis and Comparison of PID and State Feedback Controller in Inverted Pendulum System

• Main Authors: Kuo-Chun Wu, 吳國駿
• Other Authors: Lin Hong
• Format: Others
• Language: zh-TW
• Published: 2007
• Online Access: http://ndltd.ncl.edu.tw/handle/48174752050111038466

碩士 === 國立高雄應用科技大學 === 電機工程系博碩士班 === 96 === Since the inverted pendulum is a non-linear, Single input multi-output (SIMO) system with the characteristics of non-minimum phase and instability, it is widely used by scholars and researchers for verifying the performance of various controllers. This paper employs the inverted pendulum system as a control target (plant) and use PID (proportional-integral-derivative) controllers, pole placement design and linear quadratic regulator (LQR) methods, to design an effective controller. The control performance of difference controllers are compared and analyzed. The contents of this paper include the derivation of the mathematical model for the control target and depict some fundamental theories of those mentioned controllers. In Chapter 5, Matlab and Simulink software are used for simulation, and for a comparison and analysis of the control performance of each controller. In the controller designing process, we have not only to linearize the originally non-linear controlled plant but also to transform the representation model for each controller. For example, when making the pole placement design and LQR, the controlled plant has to be represented in state space form. Moreover, after a comparison and analysis on the control performance of each controller, we find that the LQR method has the best control performance among these three controllers. However, when the inverted pendulum system was simultaneously controlled on both the pendulum balancing and cart positioning, the weakness in insufficient controllability over the cart position still exists. Therefore, we attempted to integrate the LQR method with the PID controller to improve it. Simulation results show the performance has been improved significantly through the control method that integrates the LQR and PID controller.