Akiyoshi Kurobe

Keio University, Carnegie Mellon University

Papers

4

Total Citations

101

H-Index

3

About

Akiyoshi Kurobe is a leading researcher in robotics and autonomous systems, whose work fundamentally advances how machines perceive and navigate their environments. His primary research areas include 3D point cloud registration, multi-modal perception, and self-supervised learning for terrain recognition. Kurobe’s most impactful contribution is **CorsNet** (80 citations), a deep neural network that revolutionized 3D point cloud registration—a critical problem for robotics and computer vision. By outperforming classical methods like ICP, CorsNet enables more accurate alignment of 3D scans, directly improving autonomous navigation and mapping. He has also pioneered self-supervised audio-visual terrain recognition, developing systems that allow ground robots to identify and discover terrain types using combined camera and microphone data without manual labels. His work on incremental terrain clustering further extends this capability, enabling robots to adapt to new environments over time. Kurobe’s research is notable for its practical impact on social robots, assistive devices, and autonomous vehicles, bridging the gap between theoretical machine learning and real-world robotic perception. His innovative use of multi-modal self-supervision is shaping the next generation of intelligent, terrain-aware mobile platforms.

Research Focus

Key Achievements

3
H-Index
4
Papers
101
Total Citations
25
Avg Citations/Paper
🏆 Most Cited Paper
CorsNet: 3D Point Cloud Registration by Deep Neural Network
80 citations · 2020
📈 Most Prolific Year: 2020 (2 Papers)
🤝 Key Collaborators: 7
🏛 Institutions: Keio University, Carnegie Mellon University

Top Papers

  1. 1
  2. 2
  3. 3
  4. 4

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 14 days ago