About

Freek Stulp is a prominent robotics researcher whose work sits at the intersection of robot learning, motor control, and reinforcement learning. He is perhaps best known for his pioneering contributions to variable impedance control, exploring how robots can adaptively modulate their mechanical stiffness to mirror the robustness and versatility of biological motor systems — work that has garnered nearly 400 citations across two foundational papers from 2010–2011. His research extensively leverages movement primitives as building blocks for complex robot behavior, with contributions spanning movement segmentation, hierarchical reinforcement learning, and associative skill memories. Stulp has also made significant strides in robot grasp learning under uncertainty, developing strategies that allow robots to handle real-world state estimation challenges. His 2013 survey on policy improvement methods — bridging reinforcement learning and evolution strategies — remains a valuable reference for the learning robotics community. More recently, Stulp contributed to the landmark Open X-Embodiment project, a large-scale collaborative effort to build generalizable robotic foundation models across diverse datasets, accumulating over 220 citations since 2023. His body of work reflects a sustained commitment to making robots more capable, adaptable, and intelligent learners.

Research Focus

Key Achievements

28
H-Index
92
Papers
2,442
Total Citations
27
Avg Citations/Paper
🏆 Most Cited Paper
Learning variable impedance control
329 citations · 2011
📈 Most Prolific Year: 2024 (9 Papers)
🤝 Key Collaborators: 313
🏛 Institutions: University of Southern California, Institut national de recherche en sciences et technologies du numérique, Deutsches Zentrum für Luft- und Raumfahrt e. V. (DLR), École Nationale Supérieure de Techniques Avancées, Département d'Informatique, Technical University of Munich

Top Papers

  1. 1
  2. 2
  3. 3
  4. 4
  5. 5
  6. 6
  7. 7
  8. 8
  9. 9
  10. 10

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 42 days ago