1P1-M01 見え情報と距離情報を用いた移動ロボットの地図生成と位置推定(移動ロボットのための視覚)

概要

論文の詳細を見る
This paper describes a method of making maps using appearance and range information, which are used for mobile robot navigation. SLAM (simultaneous localization and mapping) methods are often used for mapping. We propose a SLAM method using SIFT features appearance information. We show by experiments that using both appearance and range information improves the localization and map accuracy regardless of the amount of available SIFT features. We also propose a localization method using the generated map and appearance and range information from stereo, and evaluation its localization accuracy.
一般社団法人日本機械学会の論文
2011-05-26