期刊
ROBOTICS AND AUTONOMOUS SYSTEMS
卷 56, 期 11, 页码 901-914出版社
ELSEVIER
DOI: 10.1016/j.robot.2008.08.009
关键词
Semantic robot maps; Outdoor mapping; Support vector machine
资金
- Engineering and Physical Sciences Research Council [GR/S62215/01] Funding Source: researchfish
The ability to extract a rich set of semantic workspace labels from sensor data gathered in complex environments is a fundamental prerequisite to any form of semantic reasoning in mobile robotics. In this paper, we present all online system for the augmentation of maps of outdoor urban environments with such higher-order, semantic labels. The system employs a shallow supervised classification hierarchy to classify scene attributes, consisting of a mixture of 2D/3D geometric and visual scene information, into a range of different workspace classes. The union of classifier responses yields a rich, composite description of the local workspace. We present extensive experimental results, using two large urban data sets collected by our research platform. (C) 2008 Elsevier B.V. All rights reserved.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据