Modeling and AI that Connects the Cyber and Physical Worlds
To realize the "advanced integration of cyber and physical space" that Society 5.0 aims for, the following cycle must be repeated.
- Responding to the complex and ever-changing conditions of the real world (physical space)
- Optimizing decisions based on data collected by sensors
- Feedback to the real world (control, manipulation, action, etc.)
In order to deal with these issues, we promote R&D to create new application fields by integrating the features of modeling/simulation and AI/machine learning technologies.
For example, we aim to provide real-time feedback to the real world, dramatically improve recognition accuracy and processing speed, and solve complex real-world problems by developing the following technologies:
- Technologies to advance machine learning using prior knowledge of the subject
- Technologies for extracting knowledge from deep learning networks that can be used for modeling and simulation
- Technologies for learning model structure as well as model parameters using machine learning
R&D Projects
Full-scale R&D Project
Development of Smart Robot for Revolution of Industry (Ohnishi PJ)Engineerable AI Techniques for Practical Applications of High-Quality Machine Learning-based Systems (Ishikawa PJ)
Feasibility Study
| <Adopted in FY2018> R&D Project Title | Summary | R&D Period |
|---|---|---|
| Design of Data-driven Hierarchical Supervisor Using Formal Methods USHIO Toshimitsu（Professor, Graduate School of Engineering Science, Osaka University） |
 (PDF:193KB) |
2018.11 -2020.03 |
| Physical agent with skill-acquiring capability OHNISHI Kouhei（Professor, Global Research Institute, Keio University） |
(PDF:188KB) |
2018.11 -2020.03 |
| Development of Weather Information Infrastructure for Autonomous Urban Cities Harmonized with Nature ONISHI Ryo（Group Leader, Center for Earth Information Science and Technology, Japan Agency for Marine-Earth Science and Technology） |
(PDF:216KB) |
2018.11 -2020.03 |
| Development of a new deep-learning model that can handle both images and symbolic data and studies on its interaction with humans SUZUKI Kenji（Professor (Specially Appointed) Institute of Innovative Research, Tokyo Institute of Technology） |
(PDF:214KB) |
2018.11 -2020.03 |
| GAN-based Robot Modeling and Its Application to Autonomous Manufacturing MORIMOTO Jun（Head of Department, Brain Information Communication Research Laboratory Group, Advanced Telecommunication Research International (ATR)） |
(PDF:197KB) |
2018.11 -2020.03 |
| Building a fast optimization technique for simulation based on machine learning YAMAZAKI Keisuke（Senior Researcher, Artificial Intelligence Research Center, National Institute of Advanced Industrial Science and Technology） |
(PDF:192KB) |
2018.11 -2020.03 |
| Value Creation by Reliable Machine Learning-based Systems YOSHIOKA Nobukazu（Associate Professor, National Institute of Informatics） |
(PDF:164KB) |
2018.11 -2020.03 |
| Engineerable AI Techniques for Practical Applications of High-Quality Machine Learning-based Systems*1 ISHIKAWA Fuyuki（Associate Professor, Information Systems Architecture Science Research Division, National Institute of Informatics） |
(PDF:184KB) |
2020.04 -2020.12 |
| Smart Robot Evolved by Learning Human Skill*2 OHNISHI Kouhei（Professor, Shin-Kawasaki Frontier Research & Education Collaborative Square, Keio University） |
(PDF:239KB) |
2020.04 -2021.04 |
*1 This project is a reorganized version of "Value Creation by Reliable Machine Learning-based Systems".
*2 This project is a reorganized version of "Physical agent with skill-acquiring capability".