Kyo Kutsuzawa
Tohoku University, Saitama University, University of Tsukuba
Papers
21
Total Citations
244
H-Index
7
About
Kyo Kutsuzawa is a robotics researcher whose work spans reinforcement learning, robot manipulation, and bioinspired locomotion control. His research sits at the intersection of machine learning and physical robotics, with a particular focus on bridging the gap between simulated training environments and real-world robot deployment. Kutsuzawa's most influential contribution is his comprehensive survey on sim-to-real transfer techniques in reinforcement learning for bioinspired robots (2021, 74 citations), which has become a key reference for researchers navigating the practical challenges of deploying learned robot behaviors. Complementing this, his work on deep reinforcement learning for robotic assembly using non-diagonal stiffness matrices (45 citations) addresses the demanding problem of contact-rich manipulation, proposing novel control strategies for precision tasks. His research extends into energy-efficient locomotion through spiking neural networks (21 citations) and multimodal bipedal locomotion using passive dynamics, demonstrating a sustained interest in biologically inspired control. He has also contributed meaningfully to nonprehensile manipulation—pancake flipping, trajectory planning, and latent-space learning models—as well as soft actuator control and tool-shape estimation. Together, his body of work reflects a versatile and forward-thinking approach to making robots capable of operating intelligently and efficiently in complex, real-world environments.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
- 10