Berthold Bäuml
Scholar

Berthold Bäuml

Google Scholar ID: fjvpDsEAAAAJ
Professor of Learning AI for Dextrous Robots (AIDX), Technical University of Munich
Deep LearningDeep Reinforcement LearningHumanoid Robotics
Citations & Impact
All-time
Citations
736
 
H-index
14
 
i10-index
23
 
Publications
20
 
Co-authors
16
list available
Resume (English only)
Academic Achievements
  • The research outcomes are applied in areas such as industrial manufacturing and service robots. The lab implements its methods using the award-winning mobile humanoid robot DLR Agile Justin with its torque-controlled DLR-Hands II. They also work towards commercializing their AI methods and developing a new generation of highly dexterous yet robust robotic hands.
Research Experience
  • Works at a joint research lab between the professorship 'Learning AI for Dextrous Robots' at the Technical University of Munich (TUM) and the Institute of Robotics & Mechatronics at the German Aerospace Center (DLR). Research focuses on deep (reinforcement) learning based on first principles and simulation models, allowing zero-shot transfer to real systems without needing human demonstrations or arbitrary internet videos.
Background
  • Focused on the principle of autonomously learning AI for dextrous robotic manipulation. The aim is to develop intelligent autonomous humanoid robots that closely match human capabilities, especially in terms of dexterous manipulation with multi-fingered hands.
Miscellany
  • Contact: Email berthold.baeuml@tum.de; LinkedIn, Google Scholar, YouTube