Papers

8

Total Citations

200

H-Index

4

About

Lars Berscheid is a roboticist whose work sits at the intersection of manipulation, self-supervised learning, and industrial automation. His research focuses on enabling robots to handle complex, deformable objects and perform precise tasks without explicit object models. Berscheid’s most notable contribution is **SpeedFolding** (87 citations), where he developed an efficient bimanual system for folding garments—a long-standing challenge due to the high-dimensional state space of cloth. He also pioneered self-supervised approaches for pick-and-place (75 citations), allowing robots to learn manipulation from a single demonstrated goal state, bypassing the need for pre-defined object models. In industrial settings, Berscheid advanced safe visuo-tactile feedback policies for high-tolerance insertion tasks, addressing the risks of part breakage during real-world learning. Beyond manipulation, he created **Ruckig**, an online trajectory generation algorithm respecting jerk-limited constraints, which has been adopted in real-time robotics systems. With a publication record spanning from 2018 to 2023, Berscheid’s work consistently emphasizes practical, deployable solutions—combining theoretical rigor with real-world validation. His research has garnered over 200 citations, reflecting its impact on both academic robotics and industrial applications.

Research Focus

Key Achievements

4
H-Index
8
Papers
200
Total Citations
25
Avg Citations/Paper
🏆 Most Cited Paper
SpeedFolding: Learning Efficient Bimanual Folding of Garments
87 citations · 2022
📈 Most Prolific Year: 2022 (2 Papers)
🤝 Key Collaborators: 14
🏛 Institutions: Karlsruhe Institute of Technology, University of Göttingen

Top Papers

  1. 1
  2. 2
  3. 3
  4. 4
  5. 5
  6. 6
  7. 7
  8. 8

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 13 days ago