A review: Simultaneous localization and mapping algorithms

Simultaneous Localization and Mapping (SLAM) involves creating an environmental map based on sensor data, while concurrently keeping track of the robot’s current position. Efficient and accurate SLAM is crucial for any mobile robot to perform robust navigation. It is also the keystone for higher-lev...

全面介绍

Saved in:
书目详细资料
Main Authors: Hadji, Saif Eddine, Kazi, Suhail, Tang, Howe Hing, Mohamed Ali, Mohamed Sultan
格式: Article
出版: Penerbit UTM Press 2015
主题:
在线阅读:http://eprints.utm.my/id/eprint/57655/
http://dx.doi.org/10.11113/jt.v73.4188
标签: 添加标签
没有标签, 成为第一个标记此记录!
实物特征
总结:Simultaneous Localization and Mapping (SLAM) involves creating an environmental map based on sensor data, while concurrently keeping track of the robot’s current position. Efficient and accurate SLAM is crucial for any mobile robot to perform robust navigation. It is also the keystone for higher-level tasks such as path planning and autonomous navigation. The past two decades have seen rapid and exciting progress in solving the SLAM problem together with many compelling implementations of SLAM methods. In this paper, we will review the two common families of SLAM algorithms: Kalman filter with its variations and particle filters. This article complements other surveys in this ?eld by reviewing the representative algorithms and the state-of-the-art in each family. It clearly identifies the inherent relationship between the state estimation via the KF versus PF techniques, all of which are derivations of Bayes rule.