Modeling and MPC-based balance control for a wheeled bipedal robot

Abstract

In recent years, wheeled bipedal robots have been widely researched as its flexible structures and superior performances. With the utilization of wheels, wheeled bipedal robots have the ability to move quickly on flat pavements. In addition to this, they can cope with some complicated terrains by motors mounted on legs. Various tasks have been achieved by wheeled bipedal robots such as trajectory tracking, carrying objects and leaping over obstacles. But whether the task is, the balance control is the basis of the subsequent movements. In this article, we established the dynamic equations of a wheeled bipedal robot on an inclined surface firstly. After linearization and discretization around the equilibrium point, a model predictive controller (MPC) with input constraints is constructed to keep the balance of the robot. In order to deal with unmodeled dynamics and external disturbances, an extended state observer (ESO) is designed to enhance the robustness of the controller. The simulation results prove that the proposed control method can ensure the balance of the robot and has a satisfactory adaptability to external interference.