PhD title: Virtual Commissioning of Industrial Machines through Bayesian Optimization

Our research group works on modelling, control and optimization for Mechatronic systems, Industrial robots and processes. We are part of the department of Electromechanical, Systems and Metal Engineering within the Faculty of Engineering of Ghent University (www.ugent.be/ea/emsme/en). Ghent University is a top 100 university worldwide and one of the major universities in Belgium, with more than 44000 students and 15000 staff members. Our campus is situated in Ghent, a lively city at the heart of Europe (visit.gent.be/en/HOME). Our research group is also associated to Flanders Make (www.flandersmake.be/en), a non-profit organization, funding precompetitive research for the betterment of the Flemish manufacturing industry. The candidate will be directly embedded in an international research group, working together as a team and will have the possibility to collaborate with many other people active at Ghent University and within Flanders.
You will work on REXPEK, a Strategic Basic Research project funded by Flanders Make. The project resides in the virtual commissioning paradigm where (near) first-time-right commissioning of machines is pursued by means of model based design approaches. The key idea is to determine optimal machine settings through virtual experiments. For this concept to work, the reality gap between the virtual and physical experiments needs to be minimized. Further, in order to reduce the remaining physical experiments, each attempt must be chosen so to yields as much information as possible. The project takes a unique angle to this problem by taking into account the human as a valuable resource both as a knowledge base, a performance sensor or critique as well as a discriminator between what could be good or bad settings. The project has the overall aim to develop new methodologies to assist operators with machine commissioning and real-time tuning.
Your specific research challenge lies in the development of new Bayesian Optimization (BO) techniques. BO is a black-box optimization strategy that operates on function evaluations alone originally developed to optimize functions that could be evaluated only through physical experimentation. You will pursue BO methods that can exploit a model based prior whilst simultaneously identifying a context probability that cannot be observed directly. In the past years our research group has accumulated critical expertise and laboratory infrastructure to support this PhD.

The candidate will be expected to

• Perform described research and strive towards successful project execution.
• Develop machine learning methodologies and software (Python) for probabilistic optimization & identification.
• Present research at conferences and in journals.
• Cooperate with researchers active within the research group and outside.
• Contribute to the teaching related to modelling and optimization.

Our offer:

• A 4 years period doctoral position.
• An internationally competitive salary that corresponds to the salary scales for Doctoral Research Fellows as established by the Flemish government.
• 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 time to apply and improve your knowledge and skills on state-of-the-art machine learning, (probabilistic) modelling, system identification and numerical optimization.
• Starting date 01/10/2022.

Job profile

We are looking for a team member with a background in (probabilistic) machine learning, numerical optimization methods and experience with system identification concepts. You are quick-witted, have an appetite for the theoretical and are keen on applying and/or improving your programming skills.

• You hold a M.Sc. in electromechanical engineering or related engineering fields such as control & automation.
• You have proven experience with numerical optimization methods in machine learning or system design.
• You have proven experience in Python.
• You have experience in or understanding of artificial intelligence and (probabilistic) machine learning methods.
• You have a team player mindset, a strong personality and work in a result-oriented manner.
• You are creative and 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, containing 1 or more references and a motivation letter to dr. Tom Lefebvre (Tom.Lefebvre@ugent.be) and dr. Saeideh Khatiry Goharood (Saeideh.KhatiryGoharoodi@UGent.be) including ‘REXPEK OPTIMIZATION PHD’ in the email subject. If you pass the pre-selection, you will receive further instructions on the selection process and will be invited for an online job interview.