SEARCH
MENU

History of Robotics Research and Development of Japan2013SensingGlobal localization for Mobile Robot using Large-scale 3D Environmental map and RGB-D Camera

Ryo KurazumeKyushu University
Jeong YongjinKyushu University
Yumi IwashitaKyushu University
Tsutomu HasegawaKyushu University
A 2D laser range finder and an ultrasound sensor are very popular devices for mobile robot localization. For example, the most of modern 2D localization techniques utilize a 2D laser range finder, and robot positions are determined based on the similarity evaluation between a 2D environmental map and 2D measured range data. However, owing to the development of a 3D laser scanner in recent years, a precise 3D environmental map is becoming available with a considerably low cost. In addition, a low cost range sensor such as Microsoft Kinect or Swiss Ranger SR4000 has been available on the market. These sensors called RGB-D cameras can capture 3D range data for a wide field of view in real-time. Therefore, 3D localization using 3D range data is expected to be popular more and more in the near future. In this research, we develop a 3D localization technique using a large-scale 3D environmental map measured by a 3D laser scanner and a 3D range data captured by a RGB-D camera on a mobile robot. Conventional 3D localization techniques using 3D environmental information utilize a registration method based on point-to-point correspondence such as Iterative Closest Point (ICP) algorithm, or voxel-to-voxel correspondence such as occupancy voxel counting. However, these techniques are computationally expensive or low accuracy due to the costly nearest point calculation or the discrete voxel representation, and hard to be applied for a global localization using a large-scale environmental map. 28th RSJ Best Paper Award in 2014. T.J.Tarn Best Paper in Robotics, 2010 IEEE International Conference on Robotics and Biomimetics (ROBIO 2010).

Related Article

記事はありません。