SIA OpenIR  > 机器人学研究室
网络遥操作若干控制问题研究
其他题名The Control Problems of the Internet-Based Tele-operation System
陈宜滨1,2
导师席宁 ; 李洪谊
分类号TP273
关键词网络遥操作 事件控制 预测控制 非线性控制 能控能观性 半张量积
索取号TP273/C49/2012
页数113页
学位专业模式识别与智能系统
学位名称博士
2012-12-01
学位授予单位中国科学院沈阳自动化研究所
学位授予地点沈阳
作者部门机器人学研究室
摘要本文主要针对以下几个方面的内容进行研究: 首先,根据网络遥操作系统的任务和控制目标,提出了以基于事件的混杂半张量积模型为参考的半自主预测控制方案,实现了相应的主从端控制软件。该控制方案不仅能充分发挥人的智能优势,而且也能发挥网络遥操作移动机器人的优势,它能很好的满足人机交互控制需求,为网络遥操作系统设计奠定了基础。其次,深入研究了基于事件参考规划和控制问题。基于事件参考规划与控制系统能够解决时延所带来的问题,能够处理路径规划中的同步协调和未知环境扰动问题,但是引入事件参考后,原系统的能控能观性及其稳定性能否不变则需要进一步深入研究。本文从两个视角研究基于事件系统的能控能观性,即从基于事件的控制系统不改变原系统反馈线性化的条件、以及能控能观分布的秩条件两个方面来研究,其中分布的秩条件更能从本质上说明问题。然后,根据网络遥操作系统的数据传输特性、人机交互特性以及控制策略等所造成的混杂特性,通过分析现有建模方法的优劣,提出了混杂半张量积建模方法,建立了混杂半张量积模型,之后将其转化为切换系统,深入地研究了系统稳定性问题。最后,通过直接无源的散射理论变换,设计了一个稳定的虚拟主从手网络移动机器人操作平台,并进行了仿真验证。为了充分提高网络遥操作系统的性能,设计了一个模型预测时延补偿控制方案,该方案首先通过将网络遥操作系统的时延分解为三个部分,然后分别进行预测估计,从而估计出整体时延,接下去对基于半张量积的混杂模型进行补偿,基于此进一步设计了非线性模型预测控制,并通过实验进行了验证。
其他摘要This thesis mainly focus on the the following subjects: First, according to the Internet-based teleoperation system’s tasks and control objectives, a semi-autonomous control scheme is advanced based on event-based hybrid semi-tensor product reference model predictive control, the softwares for master/salve are also implemented. This control scheme not only takes full use of the advantages of human intelligence, but laso takes full use of the advantages of robots. This design could well meet the human-computer interaction control requirements and laid the foundation for system design. Secondly, This thesis have a depth study on planning and control for an event-based reference systems. Event-based planning and control system is able to solve the problems brought about by the delay and is able to handle the synchronizing problem and unknown environmental disturbances in the path planning, but the controllability, observability and stability is still keep the same to the original system after introducing the event reference variable should be studied carefully. In this thesis the event-based control systems’ controllability and observability are investigated from two perspective the event system, one aspects it doesn’t change the feedback linearization conditions and the other aspects it’s that the controllability and observability distribution rank condition doesn’t change. The later aspect can capture more essence about the controllability and observability problems. Then according to hybrid characteristics caused by the data transmission, the human-computer interaction and control strategies in the Internet-based teleoperation system, through analysis of the adavantages and disadavantages for the existing modeling method proposed before, a hybrid semi-tensor product modeling approach has been advanceed and model a hybrid semi-tensor product modelby this method, and then stability study is conduct carefully by converting the hybrid semi-tensor product model into a switching system. Finally, a stable bilateral control law is developed for an Internet-based virtual master-slave mobile robot platform by the direct passive scattering transformation method and some simulations are conduct to verify this method. In order to fully improve the performance of the Internet-based teleopertion system, a model predictive time delay compensation control scheme is advanced. In this scheme the Internet-based teleoperation system’s time delay is divided into three parts, and then estimate every parts respectively, finally get the overall delay estimation. The overall delay estimation is used to compensate the semi-tensor product hybrid model, based on compensated model a nonlinear model predictive controller is designed, and verified by experiment.
语种中文
产权排序1
文献类型学位论文
条目标识符http://ir.sia.cn/handle/173321/10653
专题机器人学研究室
作者单位1.中国科学院沈阳自动化研究所
2.中国科学院大学
推荐引用方式
GB/T 7714
陈宜滨. 网络遥操作若干控制问题研究[D]. 沈阳. 中国科学院沈阳自动化研究所,2012.
条目包含的文件
文件名称/大小 文献类型 版本类型 开放类型 使用许可
网络遥操作若干控制问题研究.pdf(4438KB) 开放获取使用许可请求全文
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
[陈宜滨]的文章
百度学术
百度学术中相似的文章
[陈宜滨]的文章
必应学术
必应学术中相似的文章
[陈宜滨]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。