The research project led by Professor Tanaka at Nagoya Institute of Technology, in which I participate as a Co-PI, has been selected for JST CREST funding.
Research overview
This project aims to establish a body-fusion learning framework for transferring advanced skills from skilled practitioners to robots. It combines skill extraction from multimodal measurements, including EEG, with interactive learning through body fusion, in which a human and AI jointly control a single robot. We will identify the sensorimotor information characteristic of skilled practitioners and models through which humans and AI represent each other’s bodies, then build a learning architecture that incorporates them. Through this approach, we aim to improve the quality of demonstration data and task success rates, and enable robots to acquire multiple advanced skills.
