パノラマ表現を用いた環境地図の作成

• Item type: SIG Technical Reports(1)
• 公開日: 1992-01-23
• タイトル: パノラマ表現を用いた環境地図の作成
• タイトル / 言語: en
• タイトル: Constructing Enviromental Map Using a Panoramic Representation
• 言語: jpn
• 資源タイプ / 資源タイプ識別子: http://purl.org/coar/resource_type/c_18gh
• 資源タイプ: technical report
• 著者所属: 大阪大学基礎工学部
• 著者所属: 山梨大学工学部
• 著者所属: 大阪大学基礎工学部
• 著者所属(英): en
• 著者所属(英): Faculty of Engineering Science Osaka University
• 著者所属(英): en
• 著者所属(英): Faculty of Engineering Yamanashi University
• 著者所属(英): en
• 著者所属(英): Faculty of Engineering Science Osaka University
• 著者名: 植田, 健治; 石黒, 浩; 辻, 三郎
• 著者名(英): Kenji, Ueda; Hiroshi, Ishiguro; Saburo, Tsuji
• 論文抄録 / 内容記述タイプ: Other
• 論文抄録 / 内容記述: In this paper we describe about a method to determine the robot position for fusing local maps into a global map for an unknown envirnment. We do not use internal sensor information such as acceleration rotation velocity of wheels and so on. We employ only visual information contained in a panoramic representation. The robot position is calculated in the coordinate of the global map obtained by fusing the range in formation observed from previous robot positions. The previous robot positions are already determined in the global map and we choose one of them as a base point. The new robot position is obtained by calculating the relative motion parameters (rotation derection and destance of translation) from the base point. We calculate the rotation and moving direction of the motion parameters from FOE. Then the moving distance is remained. Using those determined motion parameters as a constraint we can restrict the candidates of the robot position along a line. We estimate each candidate by matching the new local map and the global map. The moving distance is determined by selecting the best candidate according to estimation value. Using neewly determined robot position we fuse the local map into global map. We use this improved global map to next.
• 論文抄録(英) / 内容記述タイプ: Other
• 論文抄録(英) / 内容記述: In this paper, we describe about a method to determine the robot position for fusing local maps into a global map for an unknown envirnment. We do not use internal sensor information such as acceleration, rotation velocity of wheels, and so on. We employ only visual information contained in a panoramic representation. The robot position is calculated in the coordinate of the global map obtained by fusing the range in formation observed from previous robot positions. The previous robot positions are already determined in the global map, and we choose one of them as a base point. The new robot position is obtained by calculating the relative motion parameters (rotation, derection and destance of translation) from the base point. We calculate the rotation and moving direction of the motion parameters from FOE. Then the moving distance is remained. Using those determined motion parameters as a constraint, we can restrict the candidates of the robot position along a line. We estimate each candidate by matching the new local map and the global map. The moving distance is determined by selecting the best candidate according to estimation value. Using neewly determined robot position, we fuse the local map into global map. We use this improved global map to next.
• 書誌レコードID / 収録物識別子タイプ: NCID
• 書誌レコードID / 収録物識別子: AA11131797
• 書誌情報: 情報処理学会研究報告コンピュータビジョンとイメージメディア（CVIM）; 巻 1992, 号 7(1991-CVIM-076), p. 1-8, 発行日 1992-01-23
• Notice: SIG Technical Reports are nonrefereed and hence may later appear in any journals, conferences, symposia, etc.
• 出版者 / 言語: ja
• 出版者: 情報処理学会