Yoshiyuki Ohmura

The University of Tokyo

Papers

24

Total Citations

912

H-Index

14

About

Yoshiyuki Ohmura is a pioneering robotics researcher whose work spans tactile sensing, humanoid robotics, embodied cognition, and deep imitation learning for robot manipulation. He is perhaps best known for his groundbreaking development of conformable tactile sensor skins—flexible, scalable sensing systems capable of covering curved surfaces—which garnered over 313 citations and earned a patent, fundamentally advancing how robots perceive physical contact with their environment. Building on this foundation, Ohmura demonstrated that humanoid robots could perform physically demanding whole-body tasks, such as lifting 30kg objects through tactile feedback, pushing boundaries previously thought impractical in humanoid research. His earlier theoretical contributions explored embodied cognition and the philosophical connections between humanoid embodiment and theory of mind, establishing him as a thoughtful voice bridging robotics and cognitive science. In recent years, Ohmura has turned toward transformer-based deep imitation learning for dexterous dual-arm manipulation, introducing innovative approaches such as gaze-guided learning and goal-conditioned dual-action frameworks that address the complexity of long-horizon manipulation tasks involving deformable objects. Collectively, his publications have accumulated hundreds of citations across more than two decades, reflecting a sustained and versatile research career at the intersection of hardware innovation, embodied intelligence, and modern machine learning.

Research Focus

Key Achievements

14
H-Index
24
Papers
912
Total Citations
38
Avg Citations/Paper
🏆 Most Cited Paper
Conformable and scalable tactile sensor skin for curved surfaces
313 citations · 2006
📈 Most Prolific Year: 2004 (4 Papers)
🤝 Key Collaborators: 17
🏛 Institutions: The University of Tokyo

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 · 14 days ago