LQR controller design for two-wheeled robot with a movable seat

A two-wheeled robot with a movable seat is proposed in this paper, which can turn with zero-radius and whose seat can move up and down, front and back. Firstly, the overall structure and the control system of the robot is introduced; secondly, a new filtering method called RC and Kalman filtering method is designed to get the posture of the robot; then, the robot´s dynamics model is established by Lagrange´s function method and locally linearized; finally, the LQR controller is developed to make the robot walk straight on the flat ground and verified in MATLAB/Simulink. Simulation results show that the dynamic model and the LQR controller are effective.