LMI-based Control Synthesis for a Leg/Wheel Mobile Robot

Abstract

In this study, a stabilizing controller is proposed based on model predictive control for a leg/wheel mobile robot. To assure the controller’s stability, after applying extended linearization, we employ linear matrix inequality (LMI) to design parameters that guarantee stability and feasibility. Converting kinematic constraints between a body and legs into LMI forms, we realize a controller with the stability and the kinematic constraints. A numerical simulation is conducted, and it is confirmed that a robot can move on a curved surface with a small error while satisfying the terminal condition.