Working Student – Implementation and Evaluation of Reinforcement Learning Algorithms for Collaborative Robots

In the department of Cognitive Robotics at the DLR Institute of Robotics and Mechatronics, we are developing new methods for robot learning, especially for contact-based tasks using collaborative robots. Such methods enable robots to acquire skills without explicit programming. Within the Interactive Skill Learning group of the Cognitive Robotics department, our focus lies on Learning from Demonstration (LfD) and Reinforcement Learning (RL) methods.

We are looking into combining environmental constraints, LfD, and RL in order to increase the safety of the robot while learning and decrease the time required so that skills can be learned directly on the robot without using simulation. 

Tasks:

  • Supporting the integration of new learning methods on robot
  • Conducting evaluations of learning methods on robot
  • Supporting the development of user interfaces for LfD and RL

Qualifications:

  • Hands-on knowledge of the foundations of robotics, LfD and RL
  • Basic knowledge about software frameworks in robotics (e.g., ROS)
  • Strong programming skills in Python 3 (and C++)
  • Familiar with development on Linux operating systems
  • Experience with Git

The student job position is open from January 2024.
Please send your application containing a CV and a cover letter, to Mr. Abhishek Padalkar (abhishek.padalkar[ @ ]dlr.de).