This webiste is in English. For Japanese information, goto my Researchmap. The following pictures show the robots in our lab. They are rendered using a simulation/planning system I developed under the support of NEDO. A bunch of videos about the system can be found in my Youtube channel.
Research Vision
Some of my students worry that AI and robotics are advancing too fast, and that jobs will disappear in the future. I see it the other way around. Technology is not progressing fast enough. If it truly were, we would have already conquered disease, significantly extended human lifespan, and expanded into space. Faster progress does not eliminate jobs. Instead, it creates new possibilities and entirely new industries. Occupations like typists and telephone switchboard operators have vanished, yet countless new industries have emerged in their place. The same will continue to happen. With this conviction, I am actively pursuing the integration of AI and robotics. The arrival of AI has dramatically enhanced our ability to work with complex systems. Looking ahead, I believe the following directions are particularly important: System-level design and optimization that integrates multi-sensor information, robotic mechanism design, and application requirements — essentially exploring what data to use and how to use it
Adoption of large-scale predictive models (so-called world models)
Norms in robot learning (or robot ethics) and self-evolution
Constraining and refining AI outputs using conventional mathematical models
Guided by these directions, my research rests on five pillars: Learning- and planning-based generation of task motions for robots
Design of robot hands and task-specific tools
Robot sensing, including vision, force, and tactile perception
Mutual understanding between robots and humans
Design of simulation, UMI, and teleoperation systems, and the data collection they enable
This research targets a wide range of application domains, including manufacturing, scientific research, healthcare, agriculture, and mining. Beyond these, the space industry — mentioned at the outset — also presents numerous scenarios where intelligent robots should take over tasks from humans. Although I have not yet worked directly in the space domain, I see tremendous potential there. If you are interested in intelligent robotics research or its applications in any field, please do not hesitate to reach out. Let us advance human society together.
Short Bio
I am an associate professor at the School of Engineering Science, the University of Osaka, Japan. I run a lab together with Professor Kensuke Harada and Assistant Professor Takuya Kiyokawa, focusing on robotic manipulation.
Prior to the University of Osaka, I was on a tenure-track position at the Manipulation Research Group, Intelligent System Research Institute, National Institute of Advanced Industrial Science and Technology (AIST) during 2015-2017. I was affiliated with the Japan Society for the Promotion of Science (JSPS) from 2013 to 2015 and did my postdoc research at the Manipulation Lab in the Robotics Institute, Carnegie Mellon University. I received my Ph.D. in Robotics from the Graduate School of Information Science and Technology, The University of Tokyo in 2013.
I am an IEEE senior member, with affiliations in the IEEE Robotics and Automation Society and the IEEE Systems, Man, and Cybernetics Society. I am also a member of RSJ (the Robotics Society of Japan) and JSME (Japan Society of Mechanical Engineers). I serve as an Editor on the IROS Conference Paper Review Board (CPRB) and Co-Chair of the IEEE Technical Committee on Mechanisms and Manipulation (TCMoMa).
Research Journey
The research directions outlined above grew out of a path that started with motion planning. At Peking University, I worked on motion planning under the supervision of Professor Hong Liu. During my Ph.D. at the University of Tokyo, I worked with Associate Professor Rui Fukui, Professor Tomomasa Sato, and Professor Yasuo Kuniyoshi, focusing on grasping, caging, and related mechanical design problems. During my postdoc at Carnegie Mellon University, I worked with Professor Matt Mason and focused on regrasp. At AIST, I mainly worked on industrial applications of robotic manipulation. Since joining the University of Osaka, I have been working with many talented students on both theoretical and applied aspects of the research. You may find more information about my research at my YouTube channel and my GitHub repositories:
Prospective Students
If you are considering our lab, you are probably looking at the top-tier institutes working on robotic manipulation. Thank you very much for your interest. In our lab, we challenge our students to develop practical integrated hardware and software systems. We have strong connections with industry and train our students to not only publish high-quality papers but also develop systems that solve real-world problems raised by our partners. The research topics of students working with me can be found on this page.