Papers
4
Total Citations
48
H-Index
3
About
Kenta Kato is a robotics researcher whose work bridges the critical gap between theoretical optimization and practical robotic manipulation. His primary research areas include multiobjective optimization under uncertainty, robotic bin-picking, and human-robot interfaces. Kato’s most significant contribution lies in developing optimization methods that account for the messy realities of physical experiments—specifically, heteroscedastic (input-dependent) noise and unknown failure regions, which are common in real-world robotics but often ignored in simulation. His 2016 paper on multiobjective optimization under heteroscedastic noise (30 citations) provides a foundational framework for efficiently tuning robot controllers through expensive experiments. In applied robotics, Kato’s 2019 work on bin-picking (13 citations) introduced an innovative multi-gripper switching strategy that adapts to object sparseness, addressing a key challenge in warehouse automation. He also explored teleoperation interfaces, developing a 3D CG diorama system to make mobile robot control more intuitive for non-experts. While his citation counts reflect a focused, early-career impact, Kato’s contributions are notable for their practical orientation—directly tackling the noise, cost, and failure constraints that define real robotic systems, making his work essential reading for researchers in experimental robotics and Bayesian optimization.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4