面向遥操作的虚拟向导共享控制方法研究
|
摘要
利用机器人遥操作技术代替人类在太空、深海、有毒及核辐射环境下工作,可以避免人类在恶劣环境下受到伤害,降低作业成本,提高操作效率。由于工作环境未知及工作任务复杂多变,同时受传感器、控制、计算机信息技术和人工智能等关键支撑技术的制约,研制出高智能、全自主工作的机器人很难实现。目前切实可行的解决方案是将人的智能和机器人的局部自主功能有机的结合起来,即共享控制的思想,利用人的智能参与完成机器人任务规划。本文建立基于虚拟环境的人机共享控制系统,不仅可以解决机器人遥操作系统中的时延问题,而且借助于虚拟向导提供的力觉反馈实现了人的控制与机器人系统自主决策的有机融合。
首先建立了基于虚拟环境的人机共享控制系统平台。设计了虚拟环境的总体框架结构,利用Java/Java3D面向对象技术构造出虚拟场景的基本类结构,进行了虚拟环境建模。利用JNI方法设计了力反馈式主手设备与虚拟环境的交互接口,操作者可以操纵力觉反馈主手对虚拟机器人进行关节驱动,实现了本地人机交互控制。
建立了遥操作移动机器人-虚拟向导系统的模型。建模主要包括两个部分:几何模型和运动学模型。建立系统几何模型时,本文采用Pro/E建模,再利用Java3D中objLoder接口导入进行组装。对于系统的运动学模型,本文分析了机器人的正逆运动学模型以及机器人-虚拟向导相对运动学模型。运动学建模详细分析了机器人与虚拟向导之间的交互状态,是进行力觉反馈的基础。
根据机器人末端与虚拟向导交互状态,判定机器人末端位姿与虚拟向导设定位姿的偏差,提出了基于弹簧力的位置校正、姿态校正和车体移动控制(包括路径向导和避障)的力觉反馈计算模型。建立虚拟向导力反馈系统基于二端口网络的动力学模型,进行Simulink仿真,结果表明系统能够保持时延条件下的稳定性和透明性,操作者能够对机器人末端进行力和位置的跟踪,获得稳定的力觉反馈。
最后,进行了基于虚拟向导的人机交互仿真实验研究,并对实验过程中测得的数据进行分析,验证了虚拟向导的有效性。
Using teleoperation robot instead of human in space, deep sea, toxic environment and nuclear radiation not only the operators can avoid harm in harsh environment, but also can reduce operating costs, make improvement in operation efficiency. Due to changeable operation environment, complex and dynamic operational tasks and the restriction of the key supporting techniques such as sensor technology, computer and information science and artifical intelligence, the development of high intelligent and completely autonomous robot is very hard to achieve. At present time the proper solution is the combination of part autonomy and network teleoperation of the robots to realize the integration of artificial intelligence , named shared control, which use human intelligence to complete operational task planning. In this dissertation , a human-teleoperation robot shared control system based on virtual environment has been established, not only solving problem of time delay with teleoperation robot system, but also with the help of force feedback induced by virtual fixture achieving combination of human control and autonomous decision of teleoperation robot system.
A human-teleoperation robot interactive system based on virtual environment platform was established. With overall frame structure of virtual environment constructed and basic classes of virtual scene constructed by object-oriented language Java/Java3D , virtual environment model was established. Utilized JNI(Java Native Interface) method to develop an interactive software interface between haptic master device and virtual environment and operators can manipulate joints of mobile robot with manipulating master device, achieved local human-teleoperation interactive control.
Mobile teleoperation robot/virtual guide system model has been established. Modeling includes two sides: the geometric model and kinematics model. In the geometry model, this paper has used Pro/E modeling, used Java/Java3D’s objectLoder interface for introduction and assembling the model. For the kinematics model , this paper has analyzed forward and inverse kinematics model of robot and relative kinematics model for teleoperation robot/virtual guide system. Kinematics model has analyzed interactive state between teleoperation robot and virtual guide in detail and constructs the basis of force feedback.
According to interactions and position and posture’s deviation between robot end-effector and virtual guide, computational algorithm modeled as spring of force feedback based on position control , posture control and moving of vehicle control(including path virtual fixture and obstacle avoidance)respectively was proposed. Developed a dynamic model of virtual guide force feedback system based on two-port network and used MATLAB Simulink to simulate. The results showed that the system maintained ensuring its stability while improving transparency in time delay. Operators could achieve force tracing and position tracing effectively and obtain steady force feedback.
Finally, virtual fixture was applied to simulation experiments of teleoperation and analysis of test data during experiments proved the effectiveness of virtual guide.
首先建立了基于虚拟环境的人机共享控制系统平台。设计了虚拟环境的总体框架结构,利用Java/Java3D面向对象技术构造出虚拟场景的基本类结构,进行了虚拟环境建模。利用JNI方法设计了力反馈式主手设备与虚拟环境的交互接口,操作者可以操纵力觉反馈主手对虚拟机器人进行关节驱动,实现了本地人机交互控制。
建立了遥操作移动机器人-虚拟向导系统的模型。建模主要包括两个部分:几何模型和运动学模型。建立系统几何模型时,本文采用Pro/E建模,再利用Java3D中objLoder接口导入进行组装。对于系统的运动学模型,本文分析了机器人的正逆运动学模型以及机器人-虚拟向导相对运动学模型。运动学建模详细分析了机器人与虚拟向导之间的交互状态,是进行力觉反馈的基础。
根据机器人末端与虚拟向导交互状态,判定机器人末端位姿与虚拟向导设定位姿的偏差,提出了基于弹簧力的位置校正、姿态校正和车体移动控制(包括路径向导和避障)的力觉反馈计算模型。建立虚拟向导力反馈系统基于二端口网络的动力学模型,进行Simulink仿真,结果表明系统能够保持时延条件下的稳定性和透明性,操作者能够对机器人末端进行力和位置的跟踪,获得稳定的力觉反馈。
最后,进行了基于虚拟向导的人机交互仿真实验研究,并对实验过程中测得的数据进行分析,验证了虚拟向导的有效性。
Using teleoperation robot instead of human in space, deep sea, toxic environment and nuclear radiation not only the operators can avoid harm in harsh environment, but also can reduce operating costs, make improvement in operation efficiency. Due to changeable operation environment, complex and dynamic operational tasks and the restriction of the key supporting techniques such as sensor technology, computer and information science and artifical intelligence, the development of high intelligent and completely autonomous robot is very hard to achieve. At present time the proper solution is the combination of part autonomy and network teleoperation of the robots to realize the integration of artificial intelligence , named shared control, which use human intelligence to complete operational task planning. In this dissertation , a human-teleoperation robot shared control system based on virtual environment has been established, not only solving problem of time delay with teleoperation robot system, but also with the help of force feedback induced by virtual fixture achieving combination of human control and autonomous decision of teleoperation robot system.
A human-teleoperation robot interactive system based on virtual environment platform was established. With overall frame structure of virtual environment constructed and basic classes of virtual scene constructed by object-oriented language Java/Java3D , virtual environment model was established. Utilized JNI(Java Native Interface) method to develop an interactive software interface between haptic master device and virtual environment and operators can manipulate joints of mobile robot with manipulating master device, achieved local human-teleoperation interactive control.
Mobile teleoperation robot/virtual guide system model has been established. Modeling includes two sides: the geometric model and kinematics model. In the geometry model, this paper has used Pro/E modeling, used Java/Java3D’s objectLoder interface for introduction and assembling the model. For the kinematics model , this paper has analyzed forward and inverse kinematics model of robot and relative kinematics model for teleoperation robot/virtual guide system. Kinematics model has analyzed interactive state between teleoperation robot and virtual guide in detail and constructs the basis of force feedback.
According to interactions and position and posture’s deviation between robot end-effector and virtual guide, computational algorithm modeled as spring of force feedback based on position control , posture control and moving of vehicle control(including path virtual fixture and obstacle avoidance)respectively was proposed. Developed a dynamic model of virtual guide force feedback system based on two-port network and used MATLAB Simulink to simulate. The results showed that the system maintained ensuring its stability while improving transparency in time delay. Operators could achieve force tracing and position tracing effectively and obtain steady force feedback.
Finally, virtual fixture was applied to simulation experiments of teleoperation and analysis of test data during experiments proved the effectiveness of virtual guide.
引文
[1] T.B.Sheridan. Telerobotics,Automation,and Human Supervisory Control. The MIT Press, pp.1-12, 98-108, 1992.
[2] Michael Koch. Design issues for a distributed multi-user editor. Computer Supported Cooperative Work, pp. 359-378, 1995.
[3] Broll W. Distributed virtual reality for everyone - a framework for networked VR on the Internet. IEEE Computer Society Press, pp. 121-128,1997.
[4] K.Goldberg, M. Sascha, S. Genter, etc. Desktop Teleoperation Via the World Wild Web. IEEE International Conference on Robotics & Automation. 1995, Nagoya: 654-659.
[5] Rodolfo Prada and Shahram Payandeh. A Study on Design and Analysis of Virtual Fixtures for Cutting in Training Environments[C]. Eurohaptics Conference, 2005 and Symposium on Haptic Interfaces for Virtual Environment and Teleoperator Systems, pp. 375-380, 2005.
[6] Payandeh. Application of Shared Control Strategy in the Design of a Robotic Device. American Control Conference, pp. 4532-4536, 2001.
[7]闫继宏,赵楠,赵杰等.基于适应性虚拟向导的遥操作机器人系统.哈尔滨工业大学学报. 2007, 39(1): 59~63.
[8]高永生.基于Internet多机器人遥操作系统安全机制的研究.哈尔滨工业大学博士学位论文. 2007: 5-15.
[9] G. Witus, R. D. Ellis, R. Karlsen et al. Comparison of teleoperation and supervisory control for navigation and driving with degraded communications. Proceedings of Unmanned Systems Technology XII, Orlando, FL, United states, April. 2010: 1-6.
[10] T. B. Sheridan. Space teleoperation through time delay: review and prognosis. IEEE Transanctions on Robotics & Automation. 1993, 9(5): 592-606.
[11] Shahram Payandeh and Zoran Stanisic. On Application of Virtual Fixtures as an Aid Telemanipulation and Training. age Help Haptic Interfaces for Virtual Environment and Teleoperator Systems, pp. 18-23, 2002.
[12] L.Rosenberg. Virtual Fixtures: Perceptual Tools for Telerobotic Manipulation. Proceedings of IEEE Virtual Reality Int’l. Symposium, pp. 76-82, 1993.
[13] S. Otmane, M. Mallem, A. Kheddar, F. Chavand. Active Virtual Guides as an Apparatus for Augmented Reality BasedTelemanipulation System on the Internet.Simulation Symposium, 2000. (SS 2000) Proceedings. 33rd Annual, pp.185-191, 2000.
[14] S. Park, R. Howe, and D. Torchiana. Virtual fixtures for Robotic Cardiac Surgery.4th International Conference on Medical Image Computing and Computer-Assisted Intervention,Utrecht, The Netherlands, pages 14-17, 2001.
[15] Alex B. Kuang, Shahram Payandeh, Bin Zheng,Frank Henigman and Christine L. MacKenzie. Assembling Virtual Fixtures for Guidance in Training Environments. Proceedings of the 12th International Symposium on Haptic Interfaces for Virtual Environment and Teleoperator Systems . pp.367-374, 2004.
[16] Mehdi Ammi*, Antoine Ferreira?. Robotic Assisted Micromanipulation System using Virtual Fixtures and Metaphors. 2007 IEEE International Conference on Robotics and Automation Roma, Italy, 10-14 April 2007, page(s): 456-460.
[17] Panadda Marayong, Hye Sun Na, and Allison M. Okamura. Virtual Fixture Control for Compliant Human-Machine Interfaces. 2007 IEEE International Conference on Robotics and Automation Roma, Italy, 10-14 April 2007, Page(s): 4018-4024.
[18] Ankur Kapoor and Russell H. Taylor. A Constrained Optimization Approach to Virtual Fixtures for Multi-Handed Tasks. 2008 IEEE International Conference on Robotics and Automation Pasadena, CA, USA, May 19-23, 2008, Page(s): 3401-3406.
[19]蒋再男,刘宏,谢宗武,赵京东. 3D图形预测仿真及虚拟夹具的大时延遥操作技术.西安交通大学学报, 2008, 42(1): 78-81.
[20]化建宁,崔玉洁,李洪谊,王越超,席宁.具有虚拟力觉导引功能的机器人网络遥操作系统.机器人, 2010, 32(4): 523-528.
[21]张斌,黄攀峰,刘正雄,朱建军.基于虚拟夹具的交互式空间机器人遥操作实验.宇航学报, 2011, 32(2): 446-450.
[22] Mandayam A. Srinivasan and Cagatay Basdogan. Haptics in virtual environments: taxonomy, research status, and challenges[J]. Comput.& Graphics, 1997, 21(4): 393-404.
[23]王党校,张玉茹,王玉慧,吕培军,王勇.实时力觉交互中的虚拟力计算及渲染方法.系统与仿真学报, 2004, 6: 1494-1498.
[24]王爱民,戴金桥.人机交互中的力/触觉设备进展综述.工业仪表与自动化装置, 2007(2): 14-18.
[25]唐红科.基于点的触觉式曲面操作变形研究.哈尔滨工业大学硕士学位论文. 2008: 30~32.
[26] Hong Zhang Y. Daniel Liang著.孙正兴张岩蒋维等译.计算机图形学应用Java2D和3D[M].机械工业出版社, 2008.
[27]张国亮.空间机器人遥操作中基于视觉的局部自主控制研究.哈尔滨工业大学博士学位论文. 2010: 31-32.
[28]黄舟平,闫继宏,高永生,赵杰.多机器人遥操作系统分布式虚拟环境平台.机电一体化. 2009(2): 35-40.
[29] SensAble Technologies. OpenHaptics Toolkit version 2.0. Programmer’s Guide. 2005: 1-10.
[30]刘英明,李宁,张玲,朱博.基于JNI技术C++测井应用程序集成方法.石油学报. 2010, 31(6):980-984.
[31]蔡自兴.机器人学[D].清华大学出版社, 2000: 59-65.
[32] R. P. Paul. Robot manipulators: mathematics, programming, and control. Cambridge: MIT Press, 1981.
[33]熊有伦.机器人技术基础.华中科技大学出版社, 1996: 43-45.
[34]赵杰,高胜,闫继宏,高永生,蔡鹤皋.基于虚拟向导的多操作者多机器人遥操作系统.哈尔滨工业大学学报, 2005, 37(1): 5-9.
[35]蒋再男.基于虚拟现实与局部自主的空间机器人遥操作技术研究.哈尔滨工业大学博士学位论文. 2009: 36-53.
[36] C.Preusche and G.Hirzinger.Haptics in Telerobotics-Current and Future Research and Applications.Visual Computation. 2007, 23(1): 273~284.
[37] J.Rosell,C.Vazquez and etc..Motion Planning for Haptic Guidance.Journal of Intelligent Robot System. 2008, 53(1):223~245.
[38] I.Tomotaka,Y.Koji and etc.. New Predictive Display Method of Motion and Force Information for Network Teleoperation without Virtual Environmnet Model. Proceeding of IEEE Conference on Intelligent Robots and Systems. Las Vegas,United States. 2003:2815~2822.
[39] Ming Li and Russell H. Taylor. Spatial Motion Constraints in Medical Robot Using Virtual Fixtures Generated by Anatomy. Proceedings of the 2004 IEEE International Conference on Robotics&Automation. New Orleans, LA *April 2004:1270~1275.
[40]邓启文,韦庆,李泽湘.大时延力反馈遥操作系统的PID控制.宇航学报, 2006, 27(2): 196-200.
[41] Lee D and Spong M W. Passive bilateral control of teleoperators under constant time-delay. In IFAC World Congress. 2005.
[42]陈惠开,吴新余,吴叔美.现代网络分析.人民邮电出版社,北京, 1992.
[43]李宽.基于绝对稳定性的空间机器人遥操作算法研究.国防科学技术大学硕士学位论文. 2008:35-40.
[44]刘冬雨.面向卫星在轨维护的机器人臂/手遥操作的研究.哈尔滨工业大学硕士学位论文. 2010: 41-55.
[45] Sheridan. Space Teleoperation through Time Delay: Review and Prognosis. IEEE Transaction on Robotics and Automation. 1993, 9(5): 592~606
[46]刘威.基于虚拟现实的力觉临场感遥操作研究.东南大学博士论文. 2006: 42-55.
[47]何智威.基于力觉临场感的遥操作技术研究.华中科技大学硕士学位论文. 2007: 20-25.
[48]王莉娟.虚拟手术力反馈人机交互控制系统研究.上海交通大学硕士学位论文. 2010: 20-30.
[2] Michael Koch. Design issues for a distributed multi-user editor. Computer Supported Cooperative Work, pp. 359-378, 1995.
[3] Broll W. Distributed virtual reality for everyone - a framework for networked VR on the Internet. IEEE Computer Society Press, pp. 121-128,1997.
[4] K.Goldberg, M. Sascha, S. Genter, etc. Desktop Teleoperation Via the World Wild Web. IEEE International Conference on Robotics & Automation. 1995, Nagoya: 654-659.
[5] Rodolfo Prada and Shahram Payandeh. A Study on Design and Analysis of Virtual Fixtures for Cutting in Training Environments[C]. Eurohaptics Conference, 2005 and Symposium on Haptic Interfaces for Virtual Environment and Teleoperator Systems, pp. 375-380, 2005.
[6] Payandeh. Application of Shared Control Strategy in the Design of a Robotic Device. American Control Conference, pp. 4532-4536, 2001.
[7]闫继宏,赵楠,赵杰等.基于适应性虚拟向导的遥操作机器人系统.哈尔滨工业大学学报. 2007, 39(1): 59~63.
[8]高永生.基于Internet多机器人遥操作系统安全机制的研究.哈尔滨工业大学博士学位论文. 2007: 5-15.
[9] G. Witus, R. D. Ellis, R. Karlsen et al. Comparison of teleoperation and supervisory control for navigation and driving with degraded communications. Proceedings of Unmanned Systems Technology XII, Orlando, FL, United states, April. 2010: 1-6.
[10] T. B. Sheridan. Space teleoperation through time delay: review and prognosis. IEEE Transanctions on Robotics & Automation. 1993, 9(5): 592-606.
[11] Shahram Payandeh and Zoran Stanisic. On Application of Virtual Fixtures as an Aid Telemanipulation and Training. age Help Haptic Interfaces for Virtual Environment and Teleoperator Systems, pp. 18-23, 2002.
[12] L.Rosenberg. Virtual Fixtures: Perceptual Tools for Telerobotic Manipulation. Proceedings of IEEE Virtual Reality Int’l. Symposium, pp. 76-82, 1993.
[13] S. Otmane, M. Mallem, A. Kheddar, F. Chavand. Active Virtual Guides as an Apparatus for Augmented Reality BasedTelemanipulation System on the Internet.Simulation Symposium, 2000. (SS 2000) Proceedings. 33rd Annual, pp.185-191, 2000.
[14] S. Park, R. Howe, and D. Torchiana. Virtual fixtures for Robotic Cardiac Surgery.4th International Conference on Medical Image Computing and Computer-Assisted Intervention,Utrecht, The Netherlands, pages 14-17, 2001.
[15] Alex B. Kuang, Shahram Payandeh, Bin Zheng,Frank Henigman and Christine L. MacKenzie. Assembling Virtual Fixtures for Guidance in Training Environments. Proceedings of the 12th International Symposium on Haptic Interfaces for Virtual Environment and Teleoperator Systems . pp.367-374, 2004.
[16] Mehdi Ammi*, Antoine Ferreira?. Robotic Assisted Micromanipulation System using Virtual Fixtures and Metaphors. 2007 IEEE International Conference on Robotics and Automation Roma, Italy, 10-14 April 2007, page(s): 456-460.
[17] Panadda Marayong, Hye Sun Na, and Allison M. Okamura. Virtual Fixture Control for Compliant Human-Machine Interfaces. 2007 IEEE International Conference on Robotics and Automation Roma, Italy, 10-14 April 2007, Page(s): 4018-4024.
[18] Ankur Kapoor and Russell H. Taylor. A Constrained Optimization Approach to Virtual Fixtures for Multi-Handed Tasks. 2008 IEEE International Conference on Robotics and Automation Pasadena, CA, USA, May 19-23, 2008, Page(s): 3401-3406.
[19]蒋再男,刘宏,谢宗武,赵京东. 3D图形预测仿真及虚拟夹具的大时延遥操作技术.西安交通大学学报, 2008, 42(1): 78-81.
[20]化建宁,崔玉洁,李洪谊,王越超,席宁.具有虚拟力觉导引功能的机器人网络遥操作系统.机器人, 2010, 32(4): 523-528.
[21]张斌,黄攀峰,刘正雄,朱建军.基于虚拟夹具的交互式空间机器人遥操作实验.宇航学报, 2011, 32(2): 446-450.
[22] Mandayam A. Srinivasan and Cagatay Basdogan. Haptics in virtual environments: taxonomy, research status, and challenges[J]. Comput.& Graphics, 1997, 21(4): 393-404.
[23]王党校,张玉茹,王玉慧,吕培军,王勇.实时力觉交互中的虚拟力计算及渲染方法.系统与仿真学报, 2004, 6: 1494-1498.
[24]王爱民,戴金桥.人机交互中的力/触觉设备进展综述.工业仪表与自动化装置, 2007(2): 14-18.
[25]唐红科.基于点的触觉式曲面操作变形研究.哈尔滨工业大学硕士学位论文. 2008: 30~32.
[26] Hong Zhang Y. Daniel Liang著.孙正兴张岩蒋维等译.计算机图形学应用Java2D和3D[M].机械工业出版社, 2008.
[27]张国亮.空间机器人遥操作中基于视觉的局部自主控制研究.哈尔滨工业大学博士学位论文. 2010: 31-32.
[28]黄舟平,闫继宏,高永生,赵杰.多机器人遥操作系统分布式虚拟环境平台.机电一体化. 2009(2): 35-40.
[29] SensAble Technologies. OpenHaptics Toolkit version 2.0. Programmer’s Guide. 2005: 1-10.
[30]刘英明,李宁,张玲,朱博.基于JNI技术C++测井应用程序集成方法.石油学报. 2010, 31(6):980-984.
[31]蔡自兴.机器人学[D].清华大学出版社, 2000: 59-65.
[32] R. P. Paul. Robot manipulators: mathematics, programming, and control. Cambridge: MIT Press, 1981.
[33]熊有伦.机器人技术基础.华中科技大学出版社, 1996: 43-45.
[34]赵杰,高胜,闫继宏,高永生,蔡鹤皋.基于虚拟向导的多操作者多机器人遥操作系统.哈尔滨工业大学学报, 2005, 37(1): 5-9.
[35]蒋再男.基于虚拟现实与局部自主的空间机器人遥操作技术研究.哈尔滨工业大学博士学位论文. 2009: 36-53.
[36] C.Preusche and G.Hirzinger.Haptics in Telerobotics-Current and Future Research and Applications.Visual Computation. 2007, 23(1): 273~284.
[37] J.Rosell,C.Vazquez and etc..Motion Planning for Haptic Guidance.Journal of Intelligent Robot System. 2008, 53(1):223~245.
[38] I.Tomotaka,Y.Koji and etc.. New Predictive Display Method of Motion and Force Information for Network Teleoperation without Virtual Environmnet Model. Proceeding of IEEE Conference on Intelligent Robots and Systems. Las Vegas,United States. 2003:2815~2822.
[39] Ming Li and Russell H. Taylor. Spatial Motion Constraints in Medical Robot Using Virtual Fixtures Generated by Anatomy. Proceedings of the 2004 IEEE International Conference on Robotics&Automation. New Orleans, LA *April 2004:1270~1275.
[40]邓启文,韦庆,李泽湘.大时延力反馈遥操作系统的PID控制.宇航学报, 2006, 27(2): 196-200.
[41] Lee D and Spong M W. Passive bilateral control of teleoperators under constant time-delay. In IFAC World Congress. 2005.
[42]陈惠开,吴新余,吴叔美.现代网络分析.人民邮电出版社,北京, 1992.
[43]李宽.基于绝对稳定性的空间机器人遥操作算法研究.国防科学技术大学硕士学位论文. 2008:35-40.
[44]刘冬雨.面向卫星在轨维护的机器人臂/手遥操作的研究.哈尔滨工业大学硕士学位论文. 2010: 41-55.
[45] Sheridan. Space Teleoperation through Time Delay: Review and Prognosis. IEEE Transaction on Robotics and Automation. 1993, 9(5): 592~606
[46]刘威.基于虚拟现实的力觉临场感遥操作研究.东南大学博士论文. 2006: 42-55.
[47]何智威.基于力觉临场感的遥操作技术研究.华中科技大学硕士学位论文. 2007: 20-25.
[48]王莉娟.虚拟手术力反馈人机交互控制系统研究.上海交通大学硕士学位论文. 2010: 20-30.