Static and dynamic control of network based rotary inverted pendulum system

Abstract

The control of the Rotary Inverted Pendulum (RIP) is a classic control problem that is explored often as a project in control courses due to its easily developed dynamics combined with its complexity of control design. The aim of this project is to delve into the control of this system to achiev2 the control objectives which is balancing the inverted pendulum base on CAN Network Control System. This system is composed of a pendulum attached to the end of a rotary arm controlled by a motor. The motor used is a Quanser SRV-02, whose characteristics are known completely. The motor consists of a servomotor coupled with a gear-chain. The aim is to keep the motor at a particular provided angular position. However, the major disadvantages of using NCS application are the delay introduced by the network traffic and package drop-out. Many researchers have directed their works towards solving these two issues by proposing new control algorithms or optimization of scheduling methods due to the fact that advantages offered by NCS far outweigh its centralized counterpart. In this thesis, investigation of the impact of network induced delays on the step response performance aspects of state space control designs LQR for a NCS with time delays. The vector of state-feedback control gains ‘K’ is used to deal with the degrade performance of a rotary inverted pendulum and also deal with the network traffic in network control system. Thus s simulation performed to show the effectiveness of the system. Eventually the result prove that the proposed controller gains improve the performance if RIP in a network induce delay.