Tatsuya Kamijo

Tatsuya Kamijo

Bachelor student

OMRON SINIC X | The University of Tokyo

Biography

As a recent graduate with a Bachelor’s degree in March 2024, I am actively seeking a Master’s and PhD position commencing Fall 2024. My academic and research interests are centered on developing technologies that enhance robotic capabilities to achieve safe, human-level manipulation skills in dynamic environments.

Interests
  • Robot Learning
  • Manipulation
  • Visuo-Tactile Learning
Education

  • BE in Mechanical Engineering, 2024

    The University of Tokyo

Publications

Tatsuya Kamijo, Ixchel G. Ramirez-Alpizar, Enrique Coronado, Gentiane Venture (2023). Tactile-based Active Inference for Force-Controlled Peg-in-Hole Insertions. arXiv, 2023.

arXiv

Tatsuya Kamijo, Tomoshi Iiyama, Yuta Oshima, Gentiane Venture, Tatsuya Matsushima, Yutaka Matsuo, Yusuke Iwasawa (2023). Tactile In-Hand Pose Estimation through Perceptual Inference. 2023 IROS Workshop on World Models and Predictive Coding in Cognitive Robotics, 2023 (also selected as a spotlight talk).

Tatsuya Kamijo, Koki Ishimoto, Tatsuya Matsushima, Yusuke Iwasawa, Yutaka Matsuo (2023). Visuotactile Learning with World Models. Proceedings of the 37th Annual Conference of the Japanese Society for Artificial Intelligence, 2023.

Experience

 
 
 
 
 
Research Assistant
Matsuo-Iwasawa Lab, The University of Tokyo
April 2024 – Present Tokyo
Leading a project on transfer learning of force-based skills leveraging vision-based robotics foundation models.
 
 
 
 
 
Research Intern
OMRON SINIC X Corp.
October 2023 – Present Tokyo
Leading a project on few-shot imitation learning for contact-rich manipulation tasks.
 
 
 
 
 
Teaching Assistant, Mono-Zemi
The University of Tokyo
May 2023 – July 2023 Tokyo
Mentored junior undergraduate students in the fundamentals of mechatronics, covering microcomputing, programming, and sensor technologies.
 
 
 
 
 
Software Engineer Intern
Excite Japan
August 2022 – September 2022 Tokyo
  • Developed fundamental functions for a voice call app using Flutter/Dart.
  • Developed a secure authentication feature for a manga app using Java, Kotlin.
 
 
 
 
 
Hardware Engineer Intern
Telexistence Inc.
August 2022 – August 2022 Tokyo

Projects

Tactile-based Active Inference for Force-Controlled Peg-in-Hole Insertions
Proposes a robust tactile insertion policy that can align the tilted peg with the hole using active inference, without the need for extensive training on large, diverse datasets.
Tactile-based Active Inference for Force-Controlled Peg-in-Hole Insertions

Contact

Please feel free to contact me!