Post-Doc: Delft University of Technology, The Netherlands

Post-Doc: Delft University of Technology, The Netherlands
The Delft Center for Systems and Control of the Delft University of
Technology, The Netherlands, announces an open position for a Postdoc position
(12 months) on the topic

Adaptive and Learning Control for Robotics
Department description
The Delft Centre for Systems and Control (DCSC) coordinates the education and
research activities in systems and control at Delft University of Technology.
The Centre’s research mission is to conduct fundamental research in systems
dynamics and control, involving dynamic modelling, advanced control theory,
optimization and signal analysis. The research is motivated by advanced technology
development in mechatronics and microsystems, sustainable industrial processes,
transportation and automotive systems, and physical imaging systems. The group
actively participates in the Dutch Institute of Systems and Control (DISC).

Job/project description
At the Delft Center for Systems and Control we are exploring new ideas about
learning control, by augmenting traditionally model-based controllers with
adaptation and learning capabilities. Ultimately, we aim to develop novel
control paradigms for autonomous mobile robots, moving away from the traditional
well-structured settings typically found in the area of industrial robotics.
Drawing analogies from nature, we find that a priori knowledge can play a fundamental
role in the learning speed of a generic learning controller. Although animal
brains are very plastic, (at the extreme level) it is the prior information
encoded in their DNA that endows them with, not only their physical morphology,
but also with the proper brain ''wiring'' that allows for learning to actually
be possible. The research subject of this particular Postdoc vacancy is to explore
and develop novel learning control frameworks that can be realized in physical
robots. To that end, our group has a number robotic platforms including wheeled
mobile robots, bi-pedal and multi-pedal legged robots and a humanoid robotic arm.

Requirement
The candidate must hold a PhD degree in control systems, machine learning, or
similar, with a strong background in mathematics. It is preferable that the
candidate has excellent knowledge and skills in programming (C++, Matlab,
Mathematica). Required are good communication skills in writing as well as oral,
in English (knowledge of Dutch is not required). The starting date is as soon
as possible.

More information: http://ping.fm/AsLFP