Affiliation and job title should automatically appear from the information that a researcher registered with researchmap.
Data may be outdated or undocumented.|
When there is not a connection via the internet, data are not displayed.
Development of Smart Robot for Revolution of Industry
|Prioritized Theme:||"Modeling and AI that Connects the Cyber and Physical Worlds"|
|Project Leader:|| OHNISHI Kouhei
|R&D Period:||2021.05- (Feasibility Study: 2018.11-2021.04)|
|Project Summary:||Project Summary (PDF:376KB)|
If the skills contained in the motion of experienced worker can be extracted and transferred to a robot, the scope of automation can be expanded to include atypical tasks (i.e., tasks that flexibly adapt to changes in the characteristics of an object) that are impossible with existing industrial robot. In order to achieve the above goals, the issues that need to be resolved are as follows.
1. The robot could acquire the skills of each experienced worker (learning skills).
2. Utilizing the expertise of experienced workers contained in the acquired skills to enable the robot to perform flexible movements adapted to the work object (using skills).
3. Evolving the robot ability to respond to all situations, including the unexpected and unknown (evolving skills).
The above issues can be solved by making robots smarter through advanced integration of real haptics and AI technologies. The project will bring some key technologies in the manufacturing process appeared in smart society.
The goal of smart robotics.
Kyoto University, ATR, AIST, Yokohama National University, NAIST, NEC
* This project is launched by integrating the following feasibility studies;
"GAN-based Robot Modeling and Its Application to Autonomous Manufacturing" (MORIMOTO Jun), "Building a fast optimization technique for simulation based on machine learning" (YAMAZAKI Keisuke) and "Smart Robot Evolved by Learning Human Skill" (OHNISHI Kouhei)