Context-Adaptive Intelligent Mechanism of Personalized Robots Based on Learning of Cognitive Behavioral Experiences

2020 Fiscal Year Final Research Report

• Project/Area Number: 16H01748
• Research Category: Grant-in-Aid for Scientific Research (A)
• Allocation Type: Single-year Grants
• Section: 一般
• Research Field: Intelligent robotics
• Research Institution: The University of Tokyo
• Principal Investigator: Okada Kei 東京大学, 大学院情報理工学系研究科, 教授 (70359652)
• Project Period (FY): 2016-04-01 – 2021-03-31
• Keywords: 知能ロボット / 認識行動経験学習 / 個人化学習 / パーソナライズドロボット / ロボットシステム

Outline of Final Research Achievements

The purpose of this study is to elucidate the intelligent body technology of context-adaptive service robots that generate supportive behaviors for various individuals and situations. The system consists of a mechanism for long-term experience accumulation in daily life support robots, a task materialization mechanism based on environment-independent situation-level task description and local rationality, adaptive support service task generation based on shared experience and knowledge between different robots and different environments, and service support task goal generation and serving support service experiments based on personal preferences. In this paper, we describe a system configuration method that consists of adaptive support service task generation based on reasoning, service support task goal generation based on personal preferences, and serving support service experiments.

Free Research Field

知能ロボット

Academic Significance and Societal Importance of the Research Achievements

現在，ロボットの活躍現場は工場から家庭へと展開が期待されているが，そこでは従来型の画一的なサービス支援タスク行動ではなく，その家庭独自のルールや，利用者の個人の嗜好に合わせたタスク行動の生成が必要になる．本研究は，このようなパーソナライズド・ロボットの概念を提案し，その具体的なシステム構成要素について実証的に示し，その有効性を評価している．今後，家庭で利用されるロボットに必要不可欠なシステムプロトタイプを示しており，ロボットと共生する社会の実現に貢献している．