| 非结构化场景下移动机器人FastSLAM应用研究 |
| |
| 引用本文： | 宋鑫坤,陈万米,徐昱琳,张雷,朱明.非结构化场景下移动机器人FastSLAM应用研究[J].微机发展,2010(2):95-98,102. |
| |
| 作者姓名： | 宋鑫坤 陈万米 徐昱琳 张雷 朱明 |
| |
| 作者单位： | 上海大学机电工程与自动化学院;上海市电站自动化技术重点实验室; |
| |
| 基金项目： | 上海市重点学科建设项目(T0103) |
| |
| 摘 要： | FastSLAM算法是当前移动机器人自定位与自建地图领域中研究的热点和关键。系统论述FastSLAM关键技术及基本理论，并设计非结构化场景进行自定位与自建地图应用研究。首先，对贝叶斯滤波理论进行了概述，得到移动机器人SLAM问题的基本贝叶斯滤波递归形式；其次，应用Rao—BlackweUised理论将状态分解为采样部分和解析部分进行讨论，得到SLAM问题的解耦形式；再次，论述了算法中序贯和综合重采样粒子滤波器；最后给出FastSLAM伪算法的整体实现过程，给出在非结构化场景下仿真结果。仿真结果表明FastSLAM算法在非结构化场景下能够有效快速地实现自定位与地图创建，当取粒子数合适时，在快速性和精确性方面都能够达到理想效果。
|
| 关 键 词： | FastSLAM 贝叶斯滤波 Rao-Blackwellised分解 粒子滤波器 非结构化场景 |
Study on Mobile Robot FastSLAM under Unstructured Scenes |
| |
| SONG Xin-kun,CHEN Wan-mi,XU Yu-lin,ZHANG Lei,ZHU Ming.Study on Mobile Robot FastSLAM under Unstructured Scenes[J].Microcomputer Development,2010(2):95-98,102. |
| |
| Authors: | SONG Xin-kun CHEN Wan-mi XU Yu-lin ZHANG Lei ZHU Ming |
| |
| Affiliation: | SONG Xin-kun1,CHEN Wan-mi1,2,XU Yu-lin1,ZHANG Lei1,ZHU Ming1(1.School of Electromechanical Engineering , Automation,Shanghai University,Shanghai 200072,China,2.Shanghai Key Laboratory of Power Station Automation Technology,China) |
| |
| Abstract: | In recent years,FastSLAM has been raised as hotspot in area of simultaneous localization and mapping for mobile robot.The FastSLAM key technologies are systematically summarized in this paper.First,the theoretical solution for SLAM problem based on Bayes filter is presented;Second,in order to avoid the full-state estimation,Rao-Blackwellised factorization is used to decouple the estimation process of robot and landmark state space;Third,the basic theory of particle filter is introduced,and the basic particl... |
| |
| Keywords: | FastSLAM Bayes filter Rao-Blackwellised factorization particle filter unstructured scenes |
| 本文献已被 CNKI 维普 等数据库收录！ |
|
