Register
 ALL Title Author Keyword Sponsors Type Publication date Submitted Time Subject Conference Name Source Categories KOS Subject Advisor ORCID Advanced
 SIA OpenIR  > 空间自动化技术研究室  > 学位论文
 题名: 断股补修机器人机构设计与行为规划研究 其他题名: Study on Mechanism Design and Behaviour Planning for a Broken Strand Repair Robot 作者: 宋屹峰 导师: 王洪光 分类号: TP242 关键词: 电力机器人 ; 机构设计 ; 环境状态检测 ; 行为规划 索取号: TP242/S88/2014 页码: 137页 学位专业: 机械电子工程 学位类别: 博士 答辩日期: 2014-05-29 授予单位: 中国科学院沈阳自动化研究所 作者部门: 空间自动化技术研究室 中文摘要: 随着社会发展对自动化需求的不断提高，机器人得到了巨大的发展以满足在特殊环境场合下的应用需求，能够在高空、强电磁、超高压输电线路环境完成维护检修作业的断股补修机器人正是其中的代表。特殊环境下作业的机器人既需要满足移动、越障、作业等性能需求，还要能够对复杂的外部环境状态进行检测，并完成合理的机器人行为规划，本文正是基于以上目标开展的。本文的研究工作以国家“863”计划支持项目“超高压输电线路巡检机器人技术与系统”（课题编号：2006AA04Z203）与东北电网有限公司企业项目“超高压输电线路断股压接机器人样机系统开发”（课题编号：DB-DW-SY-1009-01）为依托展开相关工作。在综述国内外现有研究热点问题的基础上，首先从机器人机构设计入手展开研究，再通过评估机器人本体各项性能完成了机器人的优化设计，而后利用视觉方法完成了输电线路的环境状态检测，并以其为基础开展了相关的机器人行为规划研究，具体研究内容如下：1.介绍了机器人机构设计并建立其运动学模型。根据维护作业任务需求，初步完成机器人机构设计、结构设计及控制系统设计。 在机构设计基础上，建立其运动学模型，并采用旋量方法完成机器人运动学正解计算。由于机器人手臂处安装有被动关节，其关节角与机器人受力状态相关，因而开展了机器人在力约束下的运动学逆解研究。通过几何解法完成力约束下的机器人运动学逆解计算，最终完成机器人的运动学研究。研究内容为后续的准静态分析、行为规划等工作奠定了基础。2.通过考察机器人性能参数开展优化设计研究。针对机器人的爬坡、越障及维护任务需求开展性能分析研究，并根据性能分析结果进行优化设计研究。爬坡性能分析着重研究由驱动轮不等半径所造成的强制滑转对牵引附着力的影响；越障性能分析利用准静态力学分析方法计算了具备被动关节结构的机器人可跨越最大障碍高度；维护性能分析计算了动力学状态下压接工具的变形状态。在以上性能分析的基础上，建立最优化目标，分别对驱动轮、移动本体及压接工具的几何参数进行优化。实验结果证明，研究内容有效的提高了机器人的整体性能水平。3.采用视觉方法进行环境状态检测研究。在对机器人采集图像进行图像裁剪、灰度化处理、边缘提取等预处理后，分别提取出图像梯度直方图与图像边缘梯度直方图作为图像特征，由于输电线正常线路、断股线路、金具障碍等环境状态在该图像特征上具备显著的差异， 采用支持向量机建立了一种基于有向图的多分类器，通过特征分类的方法完成环境检测。研究内容可以为机器人提供当前的环境状态信息，是开展机器人行为规划研究的基础。4.开展机器人的自主运动行为规划研究。 针对机器人的自主控制需求，提出了一种基于有限状态机的行为规划方法。根据检测的各类环境状态，依次针对机器人的行走、越障及维护任务建立对应的行为状态，各行为状态通过由环境状态与传感器信息建立状态决策机制来完成转移；定义机器人特定位姿状态为子状态，行走、越障及维护任务等行为状态可以通过各子状态的转移来完成。方法的有效性得到了实验验证。5.最后，在实验室与实际线路环境下依次完成了机器人的爬坡、越障、断股补修等子项功能测试实验，环境状态检测实验与行为规划实验。在此基础上，进行了综合作业实验，实验验证了文中所提方法的正确性与有效性。 英文摘要: As the requirement of automation has been constantly raised in modern society, robots are greatly developed as an advanced technological mean to meet the practical demand in special working environments.Power transmission line maintenance robot, which needs to work under the overhead, strong electromagnetic and high voltage working condition, is a representative robot among robots. Robots for special working environments need to process good performance on mobility, obstacle-crossing, maintenance manipulation, and etc. ; what is more, robots should own the capability of environment state detection and logical behaviour planning based. This thesis is carried out based on above objective.The work of this thesis is supported by two projects: 1.\emph{Extra-high Transmission Line Inspection Robot Technique and System}(No.2006AA04Z203), the National High Technical Research and Development Program of China("863" Program); 2. \emph{Extra-high Transmission Line Broken Strand Repair Robot Prototype Development}(No.DB-DW-SY-1009-01), the enterprise project of Northeast China Grid Company.After summarizing hot issues in related research field both here and abroad, the research issues of this thesis start with robot mechanical design. Thus, mechanical optimization design is carried out based on robot performance estimation. After that, a visual method is proposed for the transmission line environment state detection. At last, research on robot behaviour planning is completed based on the environment state detection. The contents of this thesis can be summarized mainly as follows.1.Robot mechanism design and kinematics. According to requirements of maintenance task, robot mechanism and control system design is initially completed. The robot kinematic modelling is set up based on the mechanism convention and forward kinematics is then solved by spinor theory. Because the passive joint installed on the robot arm is related with robot mechanical state, inverse kinematics is with force constraint. Combining quasi-static mechanical and inverse kinematics geometrical solution, robot motion synthesis is completed. The research lays foundations for the following quasi-static mechanical analysis and behaviour planning.2.Mechanical optimization design based on robot performance estimation. Robot performance analysis has been implemented aiming at climbing, obstacle-crossing and broken strand repair task, and robot mechanical optimization design is carried out based on the performance analysis. For climbing performance, research is mainly focused on the wheel traction adhesion, which is influenced by enforced wheelslip due to the varied wheel diameter. For obstacle-crossing performance, with quasi-static mechanical method, research is mainly focused on maximum height the robot with passive joints can cross. For broken strand repair performance, research is mainly focused on maintenance tool deformation in dynamical condition. Regarding above analysis as foundation, the optimization objective is set up for the geometric parameter optimization of the driving wheel, links and maintenance tool. Experiment results have demonstrated robot performance has been effectively enhanced through optimization design.3.Environment state detection by a visual method. After a series of image preprocessing, i.e. image cropping, gray processing, edge extraction and etc., histogram of gradient(HoG) and histogram of edge gradient(HoEG) are respectively extracted as the image feature. By strengthen the image gradient information in power line and normal strand edges, we can obtain the image gradient descriptor. As the HoG and HoEG descriptors show a great difference in environment states of normal wire, broken wire and counterweight, the environment state can be decided by image feature classification, where the classifier is set up by directed acyclic graph support vector machine(DAG-SVM). The detection result offers the current environment state and lays foundations for the behaviour planning.4.Robot behaviour planning based on the environment state detection. Several states are first established for walking, crossing and repairing tasks and a decision mechanism is set up by sensor information to control the transfers among all states. For each state, several special robot postures are defined as sub-states, and each task can be implemented through transferring sub-states. Experiment results have demonstrated the effectiveness of the behaviour planning method.5. Experiment. Subitem experiments are first carried out both in laboratory and live lines to test robot performance, i.e. running on a inclined line, obstacle-crossing and broken strand repairing. After that, the experiments for environment state detection and behaviour planning are also carried out. At last, a comprehensive experiment is performed to demonstrate the correctness and effectiveness of proposed menthols in this thesis. 语种: 中文 产权排序: 1 内容类型: 学位论文 URI标识: http://ir.sia.cn/handle/173321/14815 Appears in Collections: 空间自动化技术研究室_学位论文

 Files in This Item:
File Name/ File Size Content Type Version Access License

 Recommended Citation: 宋屹峰. 断股补修机器人机构设计与行为规划研究[D]. 中国科学院沈阳自动化研究所. 2014.
 Service Recommend this item Sava as my favorate item Show this item's statistics Export Endnote File Google Scholar Similar articles in Google Scholar [宋屹峰]'s Articles CSDL cross search Similar articles in CSDL Cross Search [宋屹峰]‘s Articles Related Copyright Policies Null Social Bookmarking

 您对该条目有什么异议，请填写以下表单，管理员会尽快联系您。 内 容： Email： * 单位： 验证码： 刷新
 您在IR的使用过程中有什么好的想法或者建议可以反馈给我们。 标 题： * 内 容： Email： * 验证码： 刷新