Crowdsourcing-based quick and interactive concept structure building for robots

• Project/Area Number: 23760216
• Research Category: Grant-in-Aid for Young Scientists (B)
• Allocation Type: Multi-year Fund
• Research Field: Intelligent mechanics/Mechanical systems
• Research Institution: The University of Tokyo
• Principal Investigator: WAKUDA Yuki 東京大学, 大学院・情報学環, 特任助教 (00377847)
• Project Period (FY): 2011 – 2012
• Project Status: Completed (Fiscal Year 2012)
• Budget Amount: ¥4,420,000 (Direct Cost: ¥3,400,000、Indirect Cost: ¥1,020,000); Fiscal Year 2012: ¥2,210,000 (Direct Cost: ¥1,700,000、Indirect Cost: ¥510,000); Fiscal Year 2011: ¥2,210,000 (Direct Cost: ¥1,700,000、Indirect Cost: ¥510,000)
• Keywords: 情報機器 / 知能機械システム / 機械学習 / 概念獲得 / ロボット / オントロジ / シーンアナリシス / 人工知能 / 因果律獲得 / コンピュータビジョン

Research Abstract

This study aims to achieve the visual-scene interpretation function for a real environment by a robot autonomously, particularly to efficiently optimize conceptual clustering in a concept structure building problem for real environmental event recognition. We used the crowdsource through the internet as a teaching information source of machine learning. We aim to design a part of this machine learning using the crowdsourcing approach and construct the conceptual structure of real environmental event interpretation more efficiently compared with that constructed using other general learning approaches. The achievement of this study is to design and construct the above-mentioned system, and conduct the experiment based on the constructed system.