[robotics-worldwide] open PhD position in the area of Robot Learning

dhlee at lsr.ei.tum.de dhlee at lsr.ei.tum.de
Tue Feb 8 05:36:19 PST 2011


Dynamic Human Robot Interaction for Automation System Lab, Institute of
Automatic Control Engineering, Technical University of Munich (TUM) has a
PhD opening in the research field of Machine Learning and Robot Control.

Topic: Robot Learning for Manipulation Tasks

The research aims at learning physical manipulation tasks including
constraints and dynamics information. The proposed concepts will be
implemented on a mobile robot platform with two arms and integrated into a
scenario within the DFG Cluster of Excellence, Cognition for Technical
Systems (CoTeSys).

Robot Learning by imitating humans has become an active research area in
robotics that spans across machine learning, human robot interaction, and
sensorimotor control. The available PhD research project aims at
developing a new method for robotic learning for dynamic manipulation
tasks which involves physical interaction with environment. The topic is
not limited to imitation learning but the incorporation of other machine
learning techniques is welcomed. Dynamics information is a key element for
extending the learning of human motion primitives to learning of
manipulation tasks. The robot should be enabled to learn the interaction
forces which are necessary to fulfill the demonstrated task and be able to
reproduce learned motion and force patterns. As a more complicated
problem, learning appropriate impedance behaviors can be included. It is
expected that the considered topics require a close connection between the
higher-level learning methods with the lower-level real-time controllers.

The position is open immediately. The salary is fixed and follows the
rules of the German governmental salary system. The PhD candidate will be
supervised by Prof. Dongheui Lee at TUM.

Prerequisites:

  * Master (equivalent) degree in electrical/mechanical engineering or
computer science
  * Good knowledge in robotics, control and machine learning
  * Good programming skills on C/C++ and/or Matlab simulink
  * Fluent English

Contact

If you are interested in this position, please send your application
including CV, academic transcript, relevant certificates, and selected
publications by email dhlee at tum.de. Earlier submission is advisable.

For further details, please contact dhlee at tum.de


More information about the robotics-worldwide mailing list