Bio

I am a Staff Research Scientist at Sony AI in Tokyo, working on learning and control for real-world robotic systems.

As robots move beyond structured industrial environments and into everyday settings, autonomy and learning become essential. It is no longer possible to anticipate all situations a robot may encounter after deployment, so robots must learn to act without relying on conventional programming. My work focuses on developing learning-based methods that enable robots to adapt and operate in complex, real-world environments, often involving direct physical interactions with humans.

I am particularly interested in bridging and contributing to advances in robot learning with their practical deployment on physical systems. This involves both the design of learning algorithms and their validation through experiments on real robots.

Since 2022, I have been a Senior Research Scientist at Sony AI. Previously, I worked at Preferred Networks Inc. and the ATR Computational Neuroscience Laboratories (under Jun Morimoto). I was also part of the Intelligent Autonomous Systems group at TU Darmstadt, where I worked with Jan Peters also advising students. I received my Ph.D. from the Australian Centre for Field Robotics (ACFR) under the supervision of Hugh Durrant-Whyte, Surya Singh, David Rye, and Ian Manchester. I hold a Master’s degree in control engineering from the Tokyo Institute of Technology.

Research Interests

My research focuses on enabling robots to learn and execute complex physical tasks in real-world environments. This requires integrating perception, learning, and control while ensuring safe and effective interaction with humans.

My work spans three main areas:

[Robot Learning]
Imitation and reinforcement learning on physical robots, including movement representation, probabilistic modeling, and skill acquisition from observation.

[Human–Robot Interaction and Semi-Autonomy]
Collaborative and assistive robotics, physical human–robot interaction, and interactive learning, with attention to safety, ergonomics, and shared autonomy.

[Planning and Control]
Motion planning, trajectory optimization, and adaptive control methods for robust and efficient robot behavior.