I made a physical Cart-Pole (Inverted Pendulum) system with just 272 USD and applied nonlinear model predictive control (in simulation) as well as PID control (on the physical system)!
The Cart-Pole, a classic benchmark problem in control, consists of a cart driven by a linear actuator and a pole attached via an unactuated joint. If the pole is positioned vertically, its dynamics predict it will fall due to instability. The goal of the controller is to swing the pole up and balance it above the cart. The model and controller of the inverted pendulum are foundational for many systems, including humanoid robots and SpaceX's Starship.
The process of my project can be divided into the following steps:
1. Mechanical Design: 3D modeling using Fusion 360 to design the physical structure of the system
2. Electronics: Select specifications for sensors, actuators, processors, and other components
3. Motor Control: Acceleration-based control of the stepper motor
4. Hardware Fabrication
5. Feedback (Closed) Loop: Measure real-time defined states through actuators and sensors, or estimate them using a state observer
6. Simulation Environment: Set up a simulation environment in Python to validate control strategies for the Cart-Pole system
7. Control: Formulate mathematical models and optimization problems for Model Predictive Control (MPC)
8. Experiment: Tune parameters
I won the Popular Prize at the UOS ECE Innovation Fair and presented my project to the Department of Mathematics, thanks to Professor Dohyeon Kwon’s invitation, focusing on differential equations and mathematical optimization. I received valuable advice from undergraduate students in mechanical, electrical, and computer engineering. All of it was truly enjoyable.