PhD Position in Open-World Robot Learning at University of Freiburg

Apply by June 30th, 2026
Career Stage: PhD
Freiburg

We are seeking highly motivated PhD students to join the Robot Learning Lab (https://rl.uni-freiburg.de) at University of Freiburg, headed by Prof. Abhinav Valada. The positions are fully funded (TV-L E13 100%), and the candidate will also join the ELLIS Unit Freiburg (https://ellis.uni-freiburg.de).

The PhD positions are centered on advancing open-world robot learning, particularly focusing on one or more of the directions below:

  • Open-World Perception: Developing methods that enable robots to form structured, actionable representations (e.g., scene graphs) of unseen objects, scenes, and spatial relations without relying on closed-set labels or predefined taxonomies.
  • Continual and Lifelong Adaptation: Enabling robots to learn throughout deployment, integrate new experiences, adapt to changing environments and tasks, and retain previously acquired knowledge without catastrophic forgetting.
  • Generalizable World Models: Learning predictive models of geometry, semantics, dynamics, and affordances that support reasoning, planning, and interaction in novel environments.
  • Foundation Models for Embodied Intelligence: Building multimodal robot learning models (VLA) that combine vision, language, action, and memory, supporting transfer across tasks and adaptation to new situations.
  • Open-World Decision Making and Policy Learning: Advancing reinforcement learning methods for long-horizon and hierarchical decision making and online policy adaptation, to enable robust robot behavior in the open world.

We are looking for candidates with:

  • Master's degree in computer science, robotics, machine learning, computer vision, or a related field.
  • Strong background in robotics and machine learning
  • Proficiency in Python and deep learning frameworks such as PyTorch (programming experience in C++ would be a plus)
  • Excellent written and communication skills in English

We expect outstanding qualifications, strong scientific curiosity, and clear motivation to pursue a doctoral degree. The successful candidate should be excited to conduct independent, high-impact research and contribute to ongoing projects in an interdisciplinary environment.

International candidates are strongly encouraged to apply. Knowledge of the German language is not required. Women and candidates from underrepresented groups are particularly encouraged to apply.

About the Lab:
The Robot Learning Lab is part of the ELLIS Unit Freiburg, the Department of Computer Science, and the BrainLinks-BrainTools Center. We offer an ambitious, collaborative, and international research environment at the forefront of robot learning and embodied AI. Our lab is equipped with a broad range of robotic platforms, including mobile manipulators, mobile service robots, quadrupeds, UAVs, and self-driving vehicles. We also have access to state-of-the-art compute infrastructure, including multiple GPU clusters, enabling research that spans large-scale learning, simulation, and real-world robot experiments.

About Freiburg:
Freiburg is one of Europe’s most attractive university cities, with an outstanding quality of life, a vibrant international research community, and a unique location at the edge of the Black Forest. Known for its historic center, strong environmental culture, and sunny climate, Freiburg offers an exceptional setting for both research and daily life.

How to apply:
Please send the following materials by email to rl-apply@cs.uni-freiburg.de:

  • CV
  • Motivation and research statement, including which of the above topics excites you most
  • Unofficial transcripts
  • Contact details of referees

Applications will be reviewed on a rolling basis, and the position will remain open until filled. We are excited to hear from candidates who want to help shape the future of open-world robot learning.

ELLIS Edge Newsletter
Join the 6,000+ people who get the monthly newsletter filled with the latest news, jobs, events and insights from the ELLIS Network.