SIA OpenIR  > 其他
Alternative TitleResearch on Key Technologies of Mobile Robots Navigation System
Thesis Advisor曲道奎 ; 徐方
Keyword移动机器人 导航系统 环境建模 定位 路径规划
Call NumberTP242/D79/2010
Degree Discipline模式识别与智能系统
Degree Name博士
Degree Grantor中国科学院沈阳自动化研究所
Place of Conferral沈阳
Abstract随着移动机器人应用范围的日益扩展,在动态、非结构环境下提高自主导航行为能力已经成为移动机器人研究领域的首要问题。本文以“863”高技术计划资助项目“模块化机器人体系结构”和“服务机器人”为依托,以沈阳新松机器人自动化股份有限公司自主研发的智能家用机器人为实验平台,针对移动机器人导航控制系统设计和动态环境下移动机器人导航共性方法展开深入研究,旨在提高移动机器人对动态、非结构环境的适应能力。本论文的主要内容如下: 首先,研究了基于超声传感器的栅格地图环境建模和特征地图环境建模技术。采用贝叶斯法则实现了栅格地图的环境建模和地图更新;采用Triangular-Based 方法的提取环境中的障碍物信息,并用最小二乘方法增量式提取环境的直线特征信息,实现环境的特征地图建模。 其次,针对动态非结构化未知环境下的移动机器人定位问题,设计了一种定位方法,将基于人工路标的环境地图学习和基于人工路标识别的定位过程统一起来形成完整的室内移动机器人的环境建模和定位解决方案。实验结果表明上述方法在“路标迷失”甚至在“机器人绑架”情形下移动机器人定位性能的鲁棒性。 第三,针对移动机器人动态环境下的路径规划技术进行了研究,提出了一种能够将全局路径规划方法和局部路径规划方法相结合、将基于反应的行为规划和基于慎思的行为规划相结合的路径规划解决思路,所提出的方法能够在动态环境下规划出一条全局意义最优的轨迹。 最后,探讨了移动机器人导航系统设计的相关问题。设计一种移动机器人导航系统模块化软硬件结构,并进一步研究了导航系统和人机交互系统的关系,在此基础上采用基于有限状态机模型的任务规划方法,以同时满足机器人在动态环境中的导航和交互功能。 上述研究成果已经在作者负责开发的面向家庭儿童的教育娱乐机器人、面向家庭老年人的情感陪护机器人和面向社会公共服务的智能服务机器人商业化产品上得到有效验证。
Other AbstractMobile robots are used more and more in real environments. Autonomous navigation has been one of the most critical issues for mobile robot maneuvering in dynamic and unstructured environments. This dissertation is based on “Modular Robotic Architecture” and “Service Robot” key project which are supported by the National Hi-tech Research and Development (863) Program of China. The mobile robot platform is an intelligent home robot provided by SIASUN.  In this dissertation, some of the common techniques for mobile robots working in dynamic environments are investigated to improve the robot’s ability to adapt to uncertainty environments. The main contents of this thesis are as follows: First, the gird map building and feature map building approaches based on ultrasonic sensors are studied. We use the Bayes rule to build the gird map and update it. The environment feature map is incrementally built based on extraction of the line features from obstacle points which are from the Triangular-based fusion methods. Second, to solve the localization problem in dynamic, unstructured even unknown environment, a localization method is proposed which decomposes the localization problem in dynamic unstructured unknown environment into two phases: map learning based on artificial landmark and localization based on learned map. The experiment results show the effectiveness of this algorithm, in particular, when the robot happened to be “kidnapped” or unable to see the landmark for a period of time, the proposed algorithm can still keep good localization performance. Third, a path planning strategy based on both the benefit of global path planer and local path planner and the benefit of deliberate path planner and reactive path planner is proposed for indoor mobile robot navigation in unstructured and dynamic environments. This method can not only effectively generate a global optimal path using the known information, but also handle the unknown obstacles in the dynamic environments in time. Fourth, the mobile robot system design approaches are studied. We design a modular navigation system including both hardware subsystem and software subsystem for mobile robots. Future more, we develop a task planning method based on the finite state machine principles to satisfy the requirement for mobile robots navigation and interaction in populated environments. The above results are extensively tested in three kinds of mobile robots, including the intelligent childhood educational robots, old people care robots and public service robots developed by author in SIASUN.
Contribution Rank1
Document Type学位论文
Recommended Citation
GB/T 7714
杜振军. 移动机器人导航系统关键技术研究[D]. 沈阳. 中国科学院沈阳自动化研究所,2010.
Files in This Item:
File Name/Size DocType Version Access License
移动机器人导航系统关键技术研究.pdf(2266KB) 开放获取LicenseApplication Full Text
Related Services
Recommend this item
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[杜振军]'s Articles
Baidu academic
Similar articles in Baidu academic
[杜振军]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[杜振军]'s Articles
Terms of Use
No data!
Social Bookmark/Share
All comments (0)
No comment.

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.