Automatic data-driven modeling and control for mechatronic systems

(24-05-2022) In her PhD, Saeideh Khatiry Goharoodi investigates how the process of data-driven model and controller discovery for mechatronic systems (e.g., a loom) can be automated and optimized.

'Big data' is changing our world: the way we think, work and live. It is also changing industrial developments and the research that precedes them. Driven by the ever lower cost of sensors, increasing data storage capacities and computational platforms, industry is increasingly embracing the use of this 'big data' and data-driven methods.

Researchers and engineers are therefore increasingly focused on these data-driven methods to describe and understand the physical world using mathematical equations. These data-driven approaches allow connections to be made between physical and conceptual quantities and even underlying natural laws to be found.

"Understanding the governing laws of a physical dynamic system (for example, a loom) can be used to begin to predict and control the behavior of that system. Errors can be detected, processes can be optimized increasing performance and innovations can be accelerated," Saeideh explains.

"To make this possible, mathematical models need to be built in a data-driven manner. The primary goal of my PhD is to automate the process of finding such models through the development of new algorithms."

"Through these models, we can better understand and further optimize mechatronic systems such as a loom, for example," Saeideh concludes.

Read a more detailed summary or the entire PhD


PhD Title: Data-Driven Symbolic Models for Mechatronic System Identification and Control


Contact: Saeideh Khatiry Goharoodi, Guillaume Crevecoeur, Luc Dupré

Saeideh Khatiry Goharoodi

Saeideh Khatiry born in Tehran in 1990, received her bachelor's degree in the field of mechanical engineering from Power and water university of technology in 2013. In 2015, she obtained her master's degree of mechanical engineering from Shahid Beheshti University in Tehran. Saeideh published the results of both her bachelor's and master's dissertations as conference papers.

In January 2017 she started her PhD in electromechanical engineering at Ghent university, where under supervision of Guillaume Crevecoeur and Luc Dupré, she executed research in the field of system identification and control.

In summer of 2019, Saeideh received FWO mobility grant to spend three months in Eindhoven university of technology as a guest researcher. There she had the opportunity to work together with Roland Toth who guided her to broaden her research.

Next to the research activities, Saeideh also worked on various projects from Flanders Make as a core Lab research member since January 2018.

During her PhD, Saeideh published one A1 peer-reviewed journal paper and four P1 peer-reviewed conference papers and she has a journal paper submitted.


Editor: Jeroen Ongenae - Illustrator: Roger Van Hecke