While PID and feedforward control remains the industry standard, advanced strategies like Model Predictive Control (MPC) are increasingly used in high-end applications. This project focuses on designing and optimizing a control system for a Ball-Balancing Tripod, developed as a demonstration platform to highlight the benefits of Model Predictive Control.

The tripod, a parallel manipulator with three linear actuators, stabilizes a ball with variable inertia by adjusting the platform’s tilt and height. A digital twin (DT) built in Simulink models the kinematics and dynamics and is used to compare Model Predictive Control (MPC) with a PID controller and feedforward controller. All controllers were tuned and validated on the digital twin before deployment to a Speedgoat target (Simulink Real-Time) or a Beckhoff IPC for real-time control.

The video highlights how Model Predictive Control enables faster settling times, reduce overshoot, and improve stability compared to classic control strategies.