Junji Oaki

Toshiba (Japan), Keio University

Papers

29

Total Citations

275

H-Index

9

About

Junji Oaki is a robotics researcher whose work spans robot perception, manipulation, teleoperation, and autonomous control systems. He is perhaps best known for his pioneering contributions to vision-guided robotic bin-picking, with his 2020 paper on depth image–based deep learning for grasping textureless planar-faced objects garnering 45 citations — reflecting the growing industrial demand for intelligent warehouse automation. His 2022 follow-up on learning suction graspability (22 citations) and his 2024 work on multiple-suction-cup grasping further cement his reputation as a leading voice in data-driven robotic manipulation. Beyond manipulation, Oaki has made notable contributions to telerobotics, developing multi-site teleoperation systems over ISDN networks and multi-robot collaboration frameworks that address real-world challenges like communication delay. His 2002 volleyball-playing robot demonstrated an early commitment to human-robot interaction, blending dynamic tracking with socially engaging autonomous behavior. Additional contributions include teaching-less robot systems for flexible manufacturing, elastic joint identification for serial robot arms, and control strategies for in-orbit satellite assembly. Collectively, his body of work reflects a career dedicated to bridging theoretical robotics with practical, real-world deployment across industrial, space, and collaborative settings.

Research Focus

Key Achievements

9
H-Index
29
Papers
275
Total Citations
9
Avg Citations/Paper
🏆 Most Cited Paper
Depth Image–Based Deep Learning of Grasp Planning for Textureless Planar-Faced Objects in Vision-Guided Robotic Bin-Picking
45 citations · 2020
📈 Most Prolific Year: 2003 (3 Papers)
🤝 Key Collaborators: 53
🏛 Institutions: Toshiba (Japan), Keio University

Top Papers

  1. 1
  2. 2
  3. 3
    A volleyball playing robot
    24 citations · 2002
  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