SIA OpenIR  > 机器人学研究室
天地遥操作系统的预测显示控制方法研究
Alternative TitlePredictive Display Control for Space Teleoperation System
阳方平1,2
Department机器人学研究室
Thesis Advisor王越超 ; 李洪谊
ClassificationTP273
Keyword天地遥操作 时变时延 预测显示 多目标优化 冗余空间机械臂
Call NumberTP273/Y26/2015
Pages125页
Degree Discipline模式识别与智能系统
Degree Name博士
2015-05-27
Degree Grantor中国科学院沈阳自动化研究所
Place of Conferral沈阳
Abstract天地遥操作将人的智慧延伸至太空,是空间机器人重要的操控手段之一。未来遥操作在轨服务主要包括设备状态检查、试验照料和维修维护等三类任务。由于受现有技术水平的限制,在很长一段时间内将难以研制出在非结构、复杂的空间环境中能够完全自主的空间机器人。因此仅靠空间机器人自身来完成复杂任务,不仅事实上非常困难,而且也存在巨大的安全隐患。此外,与空间站不同,在轨服务卫星是一种无人值守的航天器,不支持航天员现场作业或者在轨遥操作的方式完成任务。因此,可以发挥地面操作人员和空间机器人双方的优势,实现人机协调的天地遥操作是目前及将来很长一段时间内完成在轨服务卫星作业任务的唯一技术手段。然而在遥操作空间机器人过程中,天地通讯存在较大的时延,严重影响了系统的稳定性、透明性和操作性。因此,时延问题已成为影响空间遥操作机器人系统性能的最重要问题。 本文在分析了现有遥操作控制技术和理论的基础上,针对空间遥操作系统的大时变时延和预测显示控制方法展开了深入研究。主要研究内容包括对时延的分析和建模,以及在此基础上提出的基于预测显示的控制方法。 首先,分析天地回路时延产生因素,将各环节的时延从产生机理进行归类,采用统计学理论分析各类时延分布规律,提出分类时延到总体时延的积聚模型;并提出无源-滤波的时延环节稳定性控制方法和基于时间戳预测与PD控制器相结合的方法,对时延进行管理控制,解决时延的稳定和抖动问题。 其次,针对非接触作业任务,开展基于运动学仿真的预测显示控制方法研究。第一,提出一种避免求解雅可比矩阵伪逆的逆运动学优化算法,解决冗余机械臂的逆运动学问题;第二,提出一种多目标约束优化问题表征与求解方法,解决仿真过程中的关节空间多目标轨迹优化问题;第三,提出基于运动学仿真的预测显示控制遥操作系统框架,保证系统的稳定性与透明度。 然后,针对接触作业任务,开展基于动力学仿真的预测显示控制方法研究。提出接触目标和抓取目标动力学模型参数在线辨识方法,解决在线修正虚拟环境的几何参数和动力学参数问题。此外,提出一种对建模误差具有鲁棒性的稳定性控制算法保证系统稳定性。 最后,对天地遥操作系统地面仿真平台与实验展开研究。搭建了基于虚拟现实的天地遥操作地面仿真平台;并且,基于该平台开展实验研究,实验结果表明本文相关研究方法的有效性。 本文的工作丰富了天地遥操作系统的研究内容,针对天地遥操作系统目前存在的大时延问题提出了基于预测显示的控制方法,为后续的研究提供了理论基础和技术保障。研究成果可直接服务于中国未来空间站实验舱的科学实验和空间在轨服务,有助于我国空间技术的进步。
Other AbstractSpace teleoperation, which could extend human intelligent to space, is one of the most important methods to control space robots. Future on-orbit teleoperation services consist of equipment inspection, experiment management, maintenance and repair of spacecraft and so on. However, due to the limitations of existing robotics level, it is still difficult to develop a completely autonomous space robot under the complex, unstructured space environment. It is hard and dangerous to accomplish the complicated tasks by robot itself. Unlike space station, on-orbit service satellite is a spacecraft without human. It cannot support astronauts to accomplish tasks by field operation and on-orbit teleoperation. As a result, space teleoperaion, which could coordinate the human and machine, would be the only way to accomplish the on-orbit service tasks for a long time in the future. However, in the space teleoperation system, existed long time delay would seriously decrease the stability, transparency and operational performances. Consequently, time delay has become the prominent problem that influences the regular work of space teleoperation system. After analyzing the existing teleoperation control technologies and theories, this paper studies the issues about large time-varying delay and predictive display control method in space teleoperation systems. The main contents consist of time delay analysis and modelling, predictive display control methods for space teleoperation systems. First of all, after analyzing the factors that cause the space teleoperation time delay, we classify the time delays of different segment according to their mechanism of production. Besides, we propose an accumulation model based on the classified model using the statistical theory and method. Moreover, we introduce the passivity-filter based stability scheme and time-stamp prediction together with PD controller method to control and manage the time delay, so as to solve the stability and vibration problem of time delay. Secondly, regarding non-contact tasks, we study the kinematics simulation based predictive display control method. First, we propose an optimized inverse kinematics algorithm that avoid of computing the pseudo-inverse Jacobian matrix. Second, we introduce a formulation and solve method of constraint multi-objective optimization, so as to solve the multi-objective trajectory optimization problem in joint space. Third, we propose a predictive display teleoperation control system structure based on kinematics simulation, so as to keep the stability and transparency of system. And then, regarding contact tasks, we study the dynamic simulation based predictive display control method. First, we propose online dynamic parameters identification methods of contact object and grasp object, in order to solve the problem of on-line revising dynamic and geometry parameters of virtual environment. Second, we introduce a stability control method that is robust to modelling error, so as to keep the stability of system and solve the dynamic modelling error problem. Finally, we study ground simulation platform of space teleoperation system. First, we construct a ground simulation platform of space teleoperation system based on virtual reality. Then, we conduct experiments on this platform. The results verify the effectiveness of method proposed in this paper. Aiming at solve the large time-varying time delay issues, this work, which enriches the research of space teleoperation system, proposes a predictive display control methods for space teleoperation system. It provides a theoretical foundation and technical support for the following tasks. The research achievement could be applied to scientific experiment insdie experimental cabin of Chinese space station in future as well as space on-orbit services, and contribute to development of country’s space technology.
Language中文
Contribution Rank1
Document Type学位论文
Identifierhttp://ir.sia.cn/handle/173321/16803
Collection机器人学研究室
Affiliation1.中国科学院沈阳自动化研究所
2.中国科学院大学
Recommended Citation
GB/T 7714
阳方平. 天地遥操作系统的预测显示控制方法研究[D]. 沈阳. 中国科学院沈阳自动化研究所,2015.
Files in This Item:
File Name/Size DocType Version Access License
天地遥操作系统的预测显示控制方法研究.p(6117KB) 开放获取CC BYApplication Full Text
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[阳方平]'s Articles
Baidu academic
Similar articles in Baidu academic
[阳方平]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[阳方平]'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.