(Challenge) The main difficulty in teaching control theory is that the lecturer must not only provide all the theoretical concepts but also visualize the control system in time and frequency domains. In control system courses, the visualizations are usually provided with roughly sketches on whiteboards and thus might be difficult to understand.

 

(Solution) Dorukhan Erdem, Aykut Beke, and Assoc. Prof. Dr Tufan Kumbasar, Control and Automation Engineering Department, Istanbul Technical University have developed a deep learning based pipeline that is capable to recognize Handwritten Feedback Control Architectures on the whiteboard and to transform them into Matlab® for visualization and analysis of control systems interactively. For the working principle and real-time performance of pipeline, check https://www.youtube.com/watch?v=qQ4xG6WN-q8 

 

This study was conducted as a part of a TUBİTAK 1001 Project entitled “Development and Deployment of Deep Learning based Type-2 Fuzzy Systems” which is ending at the end of 2021.