| 应用聚类进行移动机器人定位 |
| |
| 引用本文： | 徐玉华,张崇巍,万亭亭. 应用聚类进行移动机器人定位[J]. 应用科学学报, 2009, 27(5): 532-537 |
| |
| 作者姓名： | 徐玉华 张崇巍 万亭亭 |
| |
| 作者单位： | 合肥工业大学电气与自动化工程学院，合肥230009 |
| |
| 基金项目： | 先进数控技术江苏省高校重点建设实验室基金 |
| |
| 摘 要： | 提出一种基于激光测距雷达的移动机器人定位新方法. 对每帧扫描数据进行聚类，对前后帧扫描数据按类进行匹配，获得两种匹配类，即完整匹配类和非完整匹配类. 对完整匹配类，取它们的两对端点以及质心作为匹配点；而对非完整匹配类，只取两对端点作为匹配点. 采用随机采样一致性算法剔除匹配点集中的外点，用非线性最小二乘法估计机器人运动参数，从而确定出机器人的当前位姿. 在静态和动态室内环境下进行的实验验证了该文提出方法的有效性.
|
| 关 键 词： | 移动机器人 定位 聚类 随机采样一致性算法 激光测距雷达 |
| 收稿时间： | 2009-06-12 |
| 修稿时间： | 2009-08-12 |
Localization of Mobile Robot Based on Clustering |
| |
| XU Yu-hua,ZHANG Chong-wei,WAN Ting-ting. Localization of Mobile Robot Based on Clustering[J]. Journal of Applied Sciences, 2009, 27(5): 532-537 |
| |
| Authors: | XU Yu-hua ZHANG Chong-wei WAN Ting-ting |
| |
| Affiliation: | School of Electrical Engineering and Automation, Hefei University of Technology, Hefei 230009, China |
| |
| Abstract: | In this paper, a new localization method for mobile robot is developed based on a laser range finder. The scanned data points in each frame are first divided into clusters. The current and previous scans are matchedaccording to the clusters to obtain two types of match clusters, holonomic matches and nonholonomic. For a pairof holonomically matched clusters, both pairs of endpoints and centroids are considered as match points, whilefor nonholonomically matched clusters, only endpoints are considered as match points. Random sample consensus (RANSAC) algorithm is then used to remove outliers, and nonlinear least squares method is adopted to estimate the motion parameters of the mobile robot. Experimental results demonstrate that the approach has satisfactory performance in both static and dynamic indoor environments. |
| |
| Keywords: | mobile robot localization clustering RANSAC laser range finder |
| 本文献已被 万方数据 等数据库收录！ |
| 点击此处可从《应用科学学报》浏览原始摘要信息 |
|
点击此处可从《应用科学学报》下载全文 |
