Doctoral fellow

Last application date
Jul 10, 2020 20:46
TW08 - Department of Electromechanical, Systems and Metal Engineering
Employment category
Doctoral fellow
Limited duration
M.Sc. electromechanical engineering, M.Sc. mechanical engineering, M.Sc. in physics, M.Sc. control engineering
Occupancy rate
Vacancy Type
Research staff

Job description


European factories are facing major transitions to adapt to global competitive pressures by developing the necessary key enabling technologies across a broad range of sectors. More specifically, there is an increasing global consumer demand for more customized, higher performance products. The omnipresence of data and data analytics accelerated by amongst others the decreasing cost of sensors and the ever increasing computational power can provide new initiatives to answer the global demand. Next to this is that artificial intelligence is becoming more and more mature offering real potential to be used as a relevant engineering tool in many applications. One of these applications relates to machines as can be found in the products of the European factories such as industrial machines and in robotics used in assembly lines. This trend driven by digital evolution and increasing connectivity is often referred to as Industry 4.0.

Data-driven methodologies succeed in finding strongly non-linear regression models directly from input-output data. They are able to attain performances for certain tasks that are unachievable using traditional engineering. This requires them to rely upon a large amount of data of a specific task and at the same time restricts them to that system. As the successes of these methods in the industry become apparent, how can these algorithms learn from multiple machines? Leveraging upon the data of an entire fleet of systems can result in both a faster deployment of deep learning techniques in the industry and give rise to more robust algorithms by learning from all machines’ experiences.

The project

MultiSysLeCo (Multisystems Learning Control) is a project carried out by Ghent University, KU Leuven and Flanders Make, backed by a consortium of industrial partners. The project aims to further develop current learning control techniques to enable them to leverage upon data of an entire fleet.

Within this project, UGent (PI: prof. Guillaume Crevecoeur) synthesizes data driven multisystem learning control methodologies. The methodologies can be furthermore informed and/or based on the physics in the mechatronic systems. This work will strengthen the competences already present within the group on hybrid data-driven and physics-based AI for real world mechatronic and industrial robotic applications.


Job description

The mission of the doctoral candidate is to intertwine the knowledge of present day nonlinear optimal control methodologies with advanced reinforcement learning methodologies. The candidate will further develop and apply present-day data-driven learning control techniques to realize learning control of multiple mechatronic systems.

  • You will set out a technological basis for black box multi-systems learning control, with a focus on reinforcement learning.
  • You will apply your methodologies both on setups available in the lab that need to be extended towards a fleet, and use cases put forward by the industrial partners.
  • You will present your recent work to industrial and scientific partners.
  • You will publish your results in scientific journals and present your work at international conferences.
  • You will cooperate with researchers active within the research group and outside.
  • You will be embedded in an internationally competitive research group that has a strong focus on connecting artificial intelligence with real physical dynamical systems.


We offer:

  • A 4 years period doctoral position
  • The candidate will have access to state-of-the-art tools and facilities, a network of Flemish companies active in the manufacturing industry, and the possibility to collaborate with other research groups.
  • The appropriate time to become experienced with data-driven methods, in order to further develop them and apply them later on.
  • Direct feedback from and ability to interact with an industrial consortium.
  • Cooperation on academic level with KU Leuven and Flanders Make and on application level with industrial partners, offering a broad range of connections and technical experiences.
  • The research group is situated in Ghent, a lively city at the heart of Europe (
  • Starting date: September 2020.

Profile of the candidate

Profile of the candidate

  • You hold a M.Sc. in electromechanical engineering or related
  • You have experience in or understanding of electromechanical/mechatronic/robotic systems.
  • You have experience in or understanding of artificial intelligence, modelling of dynamical systems, optimal control.
  • You have experience in or understanding of data-driven models and data analytics.
  • You have a team player mindset, a strong personality and work in a result-oriented manner.
  • You are creative, willing to work in a multidisciplinary context.
  • You are proficient in oral and written English and have strong communication skills.
  • You are willing to extend your network and able to talk on technical matters.

How to apply

Send your CV and a motivation letter to Prof. Guillaume Crevecoeur via email before Friday July 10, 2020.