Model-based design of a ballbot system controlled by mindstorms EV3

Abstract

Translating theoretical control principles into practical implementation remains a central challenge in control system engineering. Model-Based Design (MBD) has emerged as a crucial methodology to address this problem, especially when utilizing accessible platforms like LEGO Mindstorms EV3. This paper presents the MBD workflow for modeling and control of dynamic systems on the LEGO EV3 hardware, starting with DC motor position, speed control using a PID controller, and then advancing to a real-time balancing of a complex underactuated Ballbot system using an LQR controller. The MBD workflow includes system identification, model formulation, controller simulation, and tuning in MATLAB/Simulink, automatic code deployment to the EV3 brick and data acquisition for further analysis. Results show that this approach effectively reduces development effort, clarifies the link between theory and implementation, and highlights its broad applicability for systems of varying complexity, from simple motor control to advanced robotics.