Embedded learning and optimization for the next generation of smart industrial control systems

ESR10 – ETH Zurich 


PhD Position in Embedded Optimization for Learning-based Control

ETH Zurich, Switzerland

The PhD position is part of the European Training Network ‟ELO-X – Embedded Learning and Optimization for the neXt generation of smart industrial control systems”. ELO-X will recruit altogether 15 PhD fellows at 6 research universities and 5 international companies from 5 European countries, who will meet regularly during exchange visits, training events, workshops, and summer schools organized by the network.

The Intelligent Control Systems Group at ETH develops theory, algorithms and tools to make systems automatically perform challenging tasks.
The goal is to enable even safety-critical systems to leverage the potential of data and adaptation to satisfy high performance requirements in the presence of increasing complexity, uncertainty and interactions arising in modern applications. The recent success of machine learning, together with the availability of computation and sensing, motivate the use of learning techniques to address these challenges. Their wide-spread adoption in control is, however, severely limited by safety concerns when integrating learning in closed-loop decision-making. Our research addresses this problem by integrating control theory, learning and optimization. We apply the developed concepts to a range of different problems, including robotics, autonomous vehicle control, space applications, rehabilitation engineering, manufacturing, up to biomedical devices, where we mostly work in collaboration with partners from industry and academia.
The PhD will be integrated in a group with a strong focus and expertise in the domain of learning-based control. The position shall prepare the fellow for a career in advanced control engineering in industry or in academia.
PhD Project Description
PhD Position in Embedded Optimization for Learning-based Control:
The aim of this PhD position is to develop new controller designs and tailored computational methods that enable learning-based control for embedded control systems. While the performance and potential of  learning-based control has been demonstrated recently, the computational challenges associated with implementing these techniques reliably on embedded hardware with limited storage and computational power and at the computation times for fast-sampled mechatronics systems remain a key limiting factor for moving these techniques into industrial applications. Possible aspects and research directions include:

Analysis of suitable learning-based models for optimization-based control schemes (such as model predictive control) and the resulting computational properties, as well as development of tailored optimization routines.

Efficient computational methods for dual control problems, i.e. combining exploration and exploitation.

Efficient data selection and reduction and fast adaptation for learning in closed-loop control.

Learning-based control formulations that provide real-time properties by design.

Controller approximation via learning-based function approximation schemes that maintain essential controller properties.

Depending on the interests and aptitudes of the selected PhD fellow, the PhD topic can lean more in one of these directions and towards method or algorithm/computational developments.
In addition to the application in cooperation with the industry partners of the project, the institute has the opportunity to test the proposed results on demonstration platforms, such as the control of small-scale race cars

Timeline and remuneration:
The ideal start time is in spring or early summer 2021. The PhD project lasts for three to four years, and is carried out at the Institute for Dynamic Systems and Control. The PhD years include at least one longer visit – a so called “secondment” – between one and six months to another group in the ELO-X network, depending on the project needs and the scientific interests of the PhD fellows. The first year is mainly dedicated to studying and getting acquainted with the relevant state of the art in learning-based control and developing research ideas, starting from the second year the PhD then focuses on method development, application problems and publications. A fourth PhD year can be added and funds are reserved for this at ETH Zurich. The remuneration is generous and will be in line with the standards of ETH Zurich.
Candidate Profiles
Ideal candidates have a master degree in one of the following disciplines or a related field: mechanical or electrical engineering, or applied mathematics with an excellent GPA. Strong analytical skills and a strong background in dynamical systems and control is required. A good background or interest in mathematical optimization is a plus, as well as programming skills (Matlab, C/C++, python). Proficient oral and written English skills are required. The position adheres to the European policy of balanced ethnicity, age and gender.
Supervisors and Main Contacts
Supervising team at ETH Zurich:
  • Prof. Dr. Melanie Zeilinger (head of Intelligent Control Systems group);
  • Andrea Carron (senior scientist working on learning-based control for robotics applications). 
Main Contacts at the ELO-X Partner Groups which could host secondments:
  • Robert Bosch GmbH: Dr. Stefan Gering (dynamic systems control);
  • Atlas Copco: Dr. Kasper Masschaele (airtec division);
  • Politecnico di Milano: Prof. Dr. Lorenzo Fagiano (professor of automation and control engineering);
  • University of Freiburg: Prof. Moritz Diehl (professor, head of systems control and optimization laboratory), Prof. Joschka Boedecker (professor for machine learning and neurorobotics).
To apply, send an email to elo-x@imtek.uni-freiburg.de in form of one single PDF attachment containing all contents or links (any other information within the email will not be processed). Subject of your email should be: “ELO-X PhD Application – ESR 10”.

Please include, in your single PDF document, the following items in this order:

  1. A cover letter incl. statement of research interests and career goals (max. 2 pages);
  2. An academic CV;
  3. Contact details of at least two referees incl. phone numbers and emails;
  4. Your diplomas and transcript of course work and grades;
  5. Sample of technical writing (publication or thesis);
  6. Proof of English language proficiency test results.


Please send your application before January 17, 2021.


Note that your PDF will be forwarded to several people in the ELO-X institutions and that in particular all Supervisory Board members of ELO-X will have access to your application material. If you want to apply to more than one ELO-X position, please create and send separate PDFs.

Marie Curie Fellowship Eligibility Criteria in Short

To be eligible, you need to be an “Early Stage Researcher” (ESR) i.e. simultaneously fulfill the following criteria at the time of recruitment:

  1. Nationality: you may be of any nationality.
  2. Mobility: you must not have resided or carried out your main activity (work, studies, etc…) in Switzerland for more than 12 months in the 3 years immediately prior to your recruitment under the ELO-X project.
  3. Qualifications and research experience: you must be in the first 4 years of your research career after the master degree was awarded.