Post by Youssef Ayad

Designing a controller in simulation is one thing; achieving the same performance on a real system is a different challenge. I'm excited to share our Ball and Beam Control System project developed at Benha University - Benha Faculty of Engineering as part of the Design of Mechatronics course. This project involved the complete mechatronic design cycle, from mathematical modeling and control design to hardware implementation, system identification, and experimental validation. The Ball and Beam system is a classical unstable control problem in which the position of a rolling ball is regulated by continuously adjusting the beam angle through closed-loop feedback control. Following the VDI 2206 methodology, we completed the project through the full mechatronic design process: 🔹 Designed the mechanical structure and linkage mechanism using CAD tools. 🔹 Performed kinematic and dynamic analysis to derive the governing equations and understand the system dynamics. 🔹 Conducted Finite Element Analysis (FEA) to evaluate stress distribution, deformation, and structural integrity before fabrication. 🔹 Developed the mathematical model and transfer function of the system, then analyzed its open-loop response, confirming the inherent instability of the plant. 🔹 Built the complete control system block diagram in MATLAB/Simulink and designed a PID controller. Root locus and stability analyses were performed to achieve the desired closed-loop performance. 🔹 Implemented the controller on an Arduino Uno using an MG995 servo motor for beam actuation and an ultrasonic sensor for real-time ball position measurement. 🔹 Established MATLAB-Arduino serial communication for real-time monitoring, visualization, and performance evaluation. 🔹 Applied System Identification techniques in MATLAB using experimental data to obtain practical plant models and compare them with the theoretical model. One of the most valuable outcomes of this project was observing the gap between simulation and real-world implementation. While the theoretical model predicted a fast and well-damped response, the practical system was influenced by sensor noise, friction, actuator nonlinearities, structural deflection, and unmodeled dynamics. Through controller tuning, filtering techniques, and system identification, we improved the system performance and achieved stable real-time ball position control. I would like to sincerely thank Dr. Amro Shafik, PhD for his continuous guidance, valuable feedback, and support throughout the project. Special thanks to Eng. Mohamed Ashraf for his technical assistance and recommendations during the design, modeling, and implementation phases. Proud to have worked alongside my teammates: Omar Metwally Mahmoud Mohamed Ebrahim Magdy 💻 GitHub Repository: https://lnkd.in/dvNPVgxk #Mechatronics #ControlSystems #PIDControl #MATLAB #Simulink #Arduino #EmbeddedSystems #SystemIdentification #FEA #MechanicalDesign #Automation #Robotics #Engineering #BenhaUniversity