CASIA OpenIR  > 毕业生  > 博士学位论文
机器人视觉伺服控制及其实现的研究
其他题名RESEARCH ON ROBOT VISUAL SERVO CONTROL AND REALIZATION
景奉水
2002-04-01
学位类型工学博士
中文摘要本文的研究是结合“开放式模块化机器人控制器开发”和“舰船合拢数控 对中系统”这两个国家“863”机器入主题关键技术项目开展的。机器人视觉伺 服控制和多机器人协调控制都是机器人领域内的重要研究方向,具有重要理论意 义和实用价值。本文对这两个方面分别进行了研究,并重点研究了机器人视觉伺 服控制。 论文首先回顾了机器人的发展概况,接着详细讨论了在机器人视觉伺服控制 方面的研究状况,并说明了本文的研究背景和主要研究内容。然后对视觉跟踪、 Visual-Motor函数和机器人视觉伺服系统的实现等问题进行了深入研究。最后, 对船体分段对接中的多机器人协调控制问题进行了研究。 在视觉跟踪研究方面,本文提出了一种基于灰度直方图的自适应阈值图像分 割方法,成功应用到了视觉跟踪的物体识别过程中。同时对跟踪微分器进行了改 进,提出了自适应微分跟踪器和预测跟踪微分器算法,并成功应用到了视觉跟踪 的ROI位置预测过程中。 在Visual-Motor函数建立方面,本文在对模糊推理系统的万能逼近特性全 面分析的基础上,提出了一种由数据自动生成基于自适应网络模糊推理系统的 Visual—Motor函数的方法。 在视觉伺服系统的实现方面,本文提出了一种基于PC的丌放式机器人控制 器系统原型,给出了具体的实现方法。在多任务条件下,对开放式机器人控制器 的通讯和调度方法进行了研究;并针对机器人两类典型应用对通讯效率和数掘完 整性的不同需求,提出了相应的任务调度和通讯的实现方法。最后将各部分结合 起来,建成了一个基于Visua-Motor函数的机器人移动物体跟踪系统。 在多机器人协调控制方面,本文给出了基于多机器人协调的船体分段对接系 统的运动学算法、轨迹规划算法和控制策略,并对系统的精度和鲁棒性进行了分 析。这一研究成果在实际工程中已经发挥了作用,不但保证了对接精度,而且大 大提高了生产效率。
英文摘要The work of this dissertation is supported in part by two National High-tech R&D Projects: "The Design of Open Robot Controller" and "Numeral Control System for Ship Block Jointing". Researches on visual servo control and multi-robot coordination are important directions in the robotics field, not only in theory but also in practice. The dissertation is related to both of these two areas, but the focus is on the former. Firstly, the general research situation in the field of robotics is reviewed, and the study status of visual servo control of robots has been discussed in detail. Then the research background and main content of the paper are introduced. After that, the problems of visual tracking, Visual-Motor function and the realization of visual control of robot are studied thoroughly. Finally, the study of multi-robot coordination in ship-block jointing is carried out. In visual tracking, an adaptive threshold section method based on grayscale histogram is proposed and has been successfully applied to the object recognition in the process of visual tracking. In addition, the Adaptive Tracking Differentiator and the Predictive Tracking Differentiator, two modification versions of the Tracking Differentiator, are presented and applied successfully to the prediction of the ROI position in the process of visual tracking. As to visual-motor function, an automatic data-driven visual-motor function building method has been proposed, based on a thoroughly analysis of the universal approximation ability of fuzzy inference system. For the realization of visual servo control system, a prototype of PC-based open robot controller is proposed and realized. After that, the communication and scheduling approaches are discussed in the context of multiple running tasks. Then, by considering the different demands of communication efficiency and data integrality for two typical applications of the robot, the corresponding methods of task scheduling and communication are proposed. Finally, a moving object tracking system is built on the results of all researches above. As to multi-robot coordination and control, the kinematics algorithm, trajectory planning method and jointing control strategy of a multi-robot based ship block jointing system are presented. The research results have been successfully applied in practice. As a result, not only the jointing system's precision can be ensured, but also the production efficiency can be improved greatly.
关键词机器人 视觉伺服 控制器 Visual-motor函数 多机器人协调 Robot Visual Servo Controller Visual-motor Function Multi-robot Coordination
语种中文
文献类型学位论文
条目标识符http://ir.ia.ac.cn/handle/173211/5727
专题毕业生_博士学位论文
推荐引用方式
GB/T 7714
景奉水. 机器人视觉伺服控制及其实现的研究[D]. 中国科学院自动化研究所. 中国科学院研究生院,2002.
条目包含的文件
条目无相关文件。
个性服务
推荐该条目
保存到收藏夹
查看访问统计
导出为Endnote文件
谷歌学术
谷歌学术中相似的文章
[景奉水]的文章
百度学术
百度学术中相似的文章
[景奉水]的文章
必应学术
必应学术中相似的文章
[景奉水]的文章
相关权益政策
暂无数据
收藏/分享
所有评论 (0)
暂无评论
 

除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。