PROFESSOR OF DECLARATIVE METHODS FOR COGNITIVE ROBOTICS
KU Leuven Campus Diepenbeek

We welcome applicants with a wide range of expertise. Expertise in functional programming is especially appreciated.

There is a vacancy for a full-time academic position (tenured or tenure track) in the Computer Science Department at KU Leuven in the area of declarative methods for cognitive robotics. The vacant position is located at Diepenbeek Campus, which is part of the Faculty of Engineering Technology of the Science and Technology Group at KU Leuven. We are looking for internationally oriented candidates with an excellent research record and with good teaching competence in Computer Science.

The most used programming language for Robotics are C/C++ and Python: the first because one wants to get the most out of the available CPU-power, the second because of its accessibility. However, in the context of cognitive robotics, these languages have many drawbacks: software built in these languages does not scale well due to a lack of appropriate abstraction mechanisms, (formal) testing and debugging is difficult, parallelization is complex, ... Especially within artificial intelligence, declarative programming languages (both logical and functional) often enable faster, more accurate and more error-proof development of software.

Topics of interest are, amongst others:

* Development of a functional control system for robots that eliminates the shortcomings aforementioned: steps have already been taken within the international research community, but the breakthrough has not yet occurred.
* Application and deepening of functional reactive programming. This approach allows asynchronous data streams of various types of data and various sources to be processed efficiently and elegantly, something that robots must increasingly be able to do.
*Development of application software using declarative programming languages. This can be done both in the form of generic functional programming languages as well as in the form of domain-specific languages (DSL).
* Application of hybrid programming forms such as probabilistic or differentiable programming. Here, a program is partly manually coded, and partly learned. This allows seamless integration of model-based approaches with recent developments such as deep learning.

See https://www.kuleuven.be/personeel/jobsite/jobs/60003386
Feel free to get in touch if you have questions.

Tom Schrijvers