CASIA OpenIR  > 毕业生  > 硕士学位论文
力觉临场感技术及其在机器人书法中的应用研究
其他题名Research on Force Telepresence and Its Application on Robot Handwriting
毕寻
学位类型工学硕士
导师李成荣 ; 罗杨宇
2012-05-30
学位授予单位中国科学院研究生院
学位授予地点中国科学院自动化研究所
学位专业控制理论与控制工程
关键词遥操作机器人 力觉临场感 虚拟现实 模糊控制 Tele-robot Force Telepresence Virtual Reality Fuzzy Control
摘要具有力觉临场感的遥操作机器人可以向用户提供从端机器人操作环境和对象的反馈力觉信息,在接触性的作业中能够增强用户对作业过程的感知能力,有助于提高操作的质量。本文主要针对力觉临场感主从式遥操作机器人(master-slave robot with force telepresence, msRFT)的系统结构和稳定性、实时性等问题进行研究,以软笔书法机器人为代表进行实现和验证。本文主要的研究内容如下: 对msRFT的系统结构进行总体设计。使用NOVINT Falcon作为主端力反馈控制器,MOTOMAN-HP3工业机器人为从端机械臂,通过开发力反馈控制器与主端上位机接口、设计从端机械臂与上位机的通信协议、构造触力信号采集与处理模块等相关工作,完成了对msRFT实验平台的搭建。进一步将msRFT实验平台用于软笔书法机器人,实现了软笔书法的主从式操作。 利用虚拟现实技术,优化并实现了msRFT的在线和离线两种工作模式,削弱了时延对系统性能的影响。针对软笔书法机器人,建立了视觉和力觉虚拟现实模型,使操作员能够实时感知操作过程中的视觉和力觉反馈信息,增强了用户操作机器人的临场感,有效提高了用户对操作环境的控制能力。 利用遥编程技术,产生符号化的控制命令,降低了msRFT中主从机器人通信线路的负载。同时提出了从端机器人的轨迹规划方法,根据轨迹曲线特征对其进行优化,提高了msRFT的自主能力。在软笔书法机器人中,对机器人中指令目标点集进行分类和裁剪,基于动态阈值自动去除其中包含信息量较少的非必要点,有效地提升了机器人的执行效率。 设计并实现了应用虚拟现实技术的引导力控制器,根据一定的先验知识,输出相应的辅助引导力。在msRFT系统中,该控制器可以分担用户操作从端机器人的用力,减缓操作人员进行遥操作时的疲劳;在虚拟力反馈训练系统中,该控制器可以通过输出力引导用户的运动方向,更好地训练用户完成指定的操作。应用模糊逻辑控制方法,有效解决了系统负载不确定对稳定性的影响,并利用接受域加权算法抑制了控制过程中的振荡。最后,成功地将引导力控制器应用于软笔书法训练。
其他摘要Robot manipulation with force telepresence can provide the user with force feedback information from the environment operated by the slave robot. Thus, in contacting tasks, user’s perception will be enhanced to improve the quality of operations. The system structure, stability and real-time character of the master-slave robot with force telepresence(msRFT) are studied in this paper. And robot calligraphy of several applicatons are realized to test our method. The main research content is introduced as following: The system structure of msRFT is overall designed. In the system, NOVINT Falcon and MOTOMAN-HP3 respectively act as master and slave robot. The experiment platform of msRFT is build through such work as: developing the interface between Falcon and the upper computer, designing communication protocols, and implementing the module of contacting force sampling. Then, it is applied in the soft pen calligraphy robot to realize the master-slave operation with force feedback. Taking advantage of virtual reality technique, the effect of time delays on the performance of msRFT is diminished, thus online or offline mode of msRFT is implemented stably. In the soft pen calligraphy robot, virtual models of vision and force are established to simulate the environment character, so that the operator can perceive the real-time feedback information to enhance the quality of the operation. Making use of the tele-programming technique, signified control command is produced to decrease the burden of communication between master and slave robots. Meanwhile, through analysis on characteristics of trajectory curve, the movement of the slave robot is planed, to improve the autonomy of msRFT. In the soft pen calligraphy robot, the set of command points is classified and clipped to remove the unnecessary points with less information. Thus, efficiency of the robot is improved dramatically. The guiding force controller, which can generate proper assisting force according to a priori knowledge, is designed and implemented. In msRFT, the controller can share part of operating force with the user, to relieve user’s fatigue during teleoperation. In the virtual force feedback training system, the controller can provide haptic guidance for pre-designed operation training. Fuzzy logic is performed in the controller, to eliminate the effect of uncertainty of the system load on stability, and the receptive field weighted average algorithm is applied to inhibit the oscillation of...
馆藏号XWLW1766
其他标识符200928014628002
语种中文
文献类型学位论文
条目标识符http://ir.ia.ac.cn/handle/173211/7629
专题毕业生_硕士学位论文
推荐引用方式
GB/T 7714
毕寻. 力觉临场感技术及其在机器人书法中的应用研究[D]. 中国科学院自动化研究所. 中国科学院研究生院,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
CASIA_20092801462800(2861KB) 暂不开放CC BY-NC-SA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
[毕寻]的文章
百度学术
百度学术中相似的文章
[毕寻]的文章
必应学术
必应学术中相似的文章
[毕寻]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。