Modeling and Control of a Magnetic Levitator on FPGA for High-Cleanliness Robotic Applications

Abstract

This work presents the modeling and control of an axial magnetic levitator as part of an active magnetic bearing (AMB) system for SCARA robots in cleanroom environments. The design employs a rotor disk suspended by four orthogonally arranged electromagnets, eliminating mechanical contact and minimizing particle generation. A nonlinear mathematical model is developed based on Kirchhoff’s voltage law and Newton’s second law, considering gravitational and electromagnetic forces. An LQG controller is implemented on an FPGA using VHDL, enabling real-time control and fast response. MATLAB/Simulink simulations and experimental tests on a physical prototype validate the system's performance. Results show stability and high precision in axial position control, demonstrating its feasibility for low-speed, high-cleanliness applications. The study also identifies future improvement opportunities through intelligent control techniques and electromechanical optimization.