1P1-A19 センサデータのクラスタリング化による地図構築の効率化

• 江丸 貴紀

  北大

• 加畑 精一

  北大

• 星野 洋平

  北大

• 小林 幸徳

  北大

書誌事項

タイトル別名

• 1P1-A19 Efficiency Improvement of Map Construction by Clustering of Sensor Data

抄録

ICP method has been widely used in order to solve SLAM problem. Generally ICP method performs map matching by using laser range sensor (LRS). Because an amount of information that is obtained by using LRS is large, ICP method requires heavy calculations. Therefore, it is difficult to implement ICP method to small size robot that has poor computational ability. In this paper, we treat LRS information as segment in order to improve an efficiency of calculation. In our experiment, about 600 distances are obtained using the LRS. By using the proposed method, it can be reduced to dozens of segment. By using the segmented information, a time to calculate of matching is dramatically reduced without decreasing the accuracy of map matching compared to ICP method. The validity of the proposed method is confirmed by performing experiments with actual equipment.

収録刊行物

• ロボティクス・メカトロニクス講演会講演概要集

  ロボティクス・メカトロニクス講演会講演概要集 2010 (0), _1P1-A19_1-_1P1-A19_4, 2010

  一般社団法人 日本機械学会

キーワード

• SLAM
• Iterative Closest Point Method
• Laser Range Sensor

詳細情報

• CRID: 1390282680912910080
• NII論文ID: 110008741469
• DOI: 10.1299/jsmermd.2010._1p1-a19_1
• ISSN: 24243124
• Web Site: https://www.jstage.jst.go.jp/article/jsmermd/2010/0/2010__1P1-A19_1/_pdf
• 本文言語コード: ja
• データソース種別:
  • JaLC
  • Crossref
  • CiNii Articles
• 抄録ライセンスフラグ: 使用不可