Recognition and Localization of Construction Material Using Scan Matching Method

カテゴリ: 部門大会

論文No: GS17-8

グループ名: 【C】平成20年電気学会電子・情報・システム部門大会講演論文集

発行日: 2008/08/20

タイトル(英語): Recognition and Localization of Construction Material Using Scan Matching Method

著者名: 都ヒョンミン (産業技術総合研究所),金 容植(産業技術総合研究所),金奉根 (産業技術総合研究所),大場光太郎 (産業技術総合研究所),Ghang Lee(Yonsei University,Korea),Seok-Heon Yun (GyeongSang National University,Korea)

著者名(英語): Hyun Min Do (AIST),Yong-Shik Kim(AIST),Bong Keun Kim (AIST),Kohtaro Ohba(AIST),Ghang Lee(Yonsei University,Korea),Seok-Heon Yun(GyeongSang National University,Korea)

キーワード: recognition|localization|scan matching|laser range finder(LRF)|steel frame

要約(日本語): Automation and robotization has been required in construction for several decades to reduce the cost and time and to increase the productivity and safety. Especially for steel construction in high-rise building, the needs for automation is emphasized because this is inherently a very dangerous and difficult job. As one procedure of automation, the equipment like robotic crane should carry steel frames from staging area to the installed spot automatically. To do this, the position and pose of steel frames should be known to robotic crane.This paper presents a recognition and localization method for steel frame materials to remedy this problem. In the proposed method, the shape model is acquired by RF-ID technology and the loading shape of steel frame materials in staging area is recognized by laser range finder (LRF) sensors. After clustering the scanned data into several objects, scan matching method is applied to recognize each object. Experimental results are included to verify the performance of the proposed method.

PDFファイルサイズ: 4,668 Kバイト