Localization in 2D Semantic Map Using Object Recognition Information from Spherical Stereo Camera

DOI Web Site

Bibliographic Information

Other Title
  • 全天球ステレオカメラでの物体認識情報を用いた意味地図内位置姿勢推定

Description

<p>In this paper, we propose a system for localization using semantic information and distance between camera and each object from spherical stereo camera. Recently, demand for a servicing robot has been increased due to understaff for work and increasing industrial accidents in aging infrastructures. Thus, our final goal is to make general people be able to operate robots due to workforce diversification. In this paper, we aim to locate a robot more accurately by using distance information. In our method, we locate a robot from a set of object center-of-gravity points on a 2D semantic map prepared in advance and a set of corresponding object center-of-gravity points obtained from a spherical stereo camera. Through the experiments, when all matching between the point clouds was successful, a robot’s rough localization was accurate.</p>

Journal

Keywords

Details 詳細情報について

Report a problem

Back to top