Design of a Sliding Mode Controller for Two-Wheeled Balancing Robot

Abstract

Nowadays, the control of mechanical systems with fewer inputs than outputs (Under-actuated systems) has become a challenging problem for control engineers. Two-wheeled balancing robots is one of the appealing examples of this category. This type of robot contains two parallel wheels and an inverted pendulum. In this research, designing of controller have been investigated for flat surfaces. For controller design, the extract dynamics of the system has been achieved based on Kane's method. Then for the two-wheeled balancing robot, one sliding mode controller has been designed for yaw angle, and another sliding mode controller has been designed to control both position and pitch angle based on a proposed sliding surface. The main feature of the proposed controllers is that all of controllers have been designed based on the nonlinear dynamics of system. Also, considering the limits of uncertainties while designing systems, the robustness of controllers have been increased. The common problem of sliding mode control is chattering phenomenon that has been greatly reduced using saturation function instead of sign function. Simulation results comparision of the designed controller with a LQR controller, validates the effectiveness of the proposed controller.

Keywords

• Two wheeled balancing robot
• TWIP
• Kane’s method
• under-actuated system
• LQR