返回
Feature based graph-SLAM in structured environments
DOI:10.1007/s10514-014-9386-z.png)
摘要
En 中文
Introducing a priori knowledge about the latent structure of the environment in simultaneous localization and mapping (SLAM), can improve the quality and consistency results of its solutions. In this paper we describe and analyze a general framework for the detection, evaluation, incorporation and removal of structure constraints into a feature-based graph formulation of SLAM. We specifically show how including different kinds and levels of features in a hierarchical manner allows the system to easily discover new structure and why it makes more sense than other possible representations. The main algorithm in this framework follows an expectation maximization approach to iteratively infer the most probable structure and the most probable map. Experimental results show how this approach is suitable for the integration of a large variety of constraints and how our method can produce nice and consistent maps in regular environments.
Keyword:
Mobile robots
Mapping
SLAM
Structured environments
AI总结
对已上传原文的论文进行重点信息的提取,主要内容包括:简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。
期刊
IF:
4.3
论文数:
1.7K
被引数:
6.2K
机构
引用论文
Generation of Human iPSC-derived Neural Progenitor Cells (NPCs) as Drug Discovery Model for Neurological and Mitochondrial Disorders
BIO-PROTOCOL
IF0

