Comparative Analysis between Computed Torque Control, LQR Control and PID Control for a Robotic Bicycle

Abstract

The aim of this article is to make a comparison between the performance of a dynamic control and two linear controllers: PID Control and a Linear Quadratic Regulator (LQR) Control when they are designed for the balance of a Robotic Bicycle. The research is focused on determining the best controller for the robot using two different performance indexes: (i) the standard deviation of the lean angle, (ii) the root mean square value of the signal control. From the experimental test was observed that the LQR control presents a lower energy consumption and lower standard deviation of the error with respect to the PID control; however the dynamic control presents the least standard deviation of the error than the two linear controllers.