Development of motion learning method for a robot with various sensors in a real environment.

2016 Fiscal Year Final Research Report

• Project/Area Number: 26730136
• Research Category: Grant-in-Aid for Young Scientists (B)
• Allocation Type: Multi-year Fund
• Research Field: Intelligent robotics
• Research Institution: Osaka University
• Principal Investigator: Nakamura Yutaka 大阪大学, 基礎工学研究科, 招へい准教授 (70403334)
• Research Collaborator: OKADOME Yuya
• Project Period (FY): 2014-04-01 – 2017-03-31
• Keywords: 知能ロボット / 多様なセンサ / 身体的インタラクション

Outline of Final Research Achievements

In order for a robot to operate in a real environment, it is inevitable to physically interact with diverse objects including human. Such a robot would face to diverse situation, and it is necessary to integrate various sensory input and take appropriate action based on the observation. For this aim, I developed a reinforcement learning method for a robot where the required sensory input is extracted from multi-modal sensory input and a sampling based roadmap method which considers the confidence of the modelling. I also developed a robot socially interact with the human in a real environment.

Free Research Field

知能ロボティクス