Multi-dimensional information recording/presentaion and Model-based Reinforcement Learning for Motor Learning/Assistance

2014 Fiscal Year Final Research Report

• Project/Area Number: 23240028
• Research Category: Grant-in-Aid for Scientific Research (A)
• Allocation Type: Single-year Grants
• Section: 一般
• Research Field: Perception information processing/Intelligent robotics
• Research Institution: Kyushu Institute of Technology (2012-2014); Nara Institute of Science and Technology (2011)
• Principal Investigator: SHIBATA Tomohiro 九州工業大学, 生命体工学研究科(研究院), 教授 (40359873)
• Co-Investigator(Kenkyū-buntansha): TAMEI Tomoya 奈良先端科学技術大学院大学, 情報科学研究科, 助教 (40548434) ; OKADA Yohei 畿央大学, 健康科学部, 助教 (80511568) ; WADA Yoshiro 奈良県立医科大学, 医学部, 講師 (80240810) ; KOMACHI Mamoru 首都大学東京, システムデザイン研究科, 准教授 (60581329)
• Project Period (FY): 2011-04-01 – 2015-03-31
• Keywords: ロボティクス / 人間機械系 / 強化学習 / 生体計測 / 運動学習支援

Outline of Final Research Achievements

We showed a reinforcement-learning agent incorporating the Asssit-As-Needed principle can accelerate motor leanring by its individual adaptivity. We also developed an In-home posture evaluation and visual-feedback training system for patients having postural deformities, and applied for dysequilibrim and Parkinson's disease. In particular, we were able to show a feasibility of the system for in-home postural rehabilitation and its immediate but short term effectivity.

Free Research Field

知能ロボティクス