气动肌肉驱动特性研究及其嵌入式控制器的设计
|
摘要
气动肌肉以其类似生物的驱动特征、良好的顺应性和安全性,使它在康复医疗设备的研制中受到了极大的关注。本文以应用在上肢康复训练机器人当中的气动肌肉驱动器作为研究对象,对气动肌肉的驱动特性、气动肌肉对拉驱动的关节刚度调节和位置控制等问题进行研究和探讨,并利用ARM最新架构的CORTEX M3微控制器搭建了适用于康复训练机器人的嵌入式控制系统。
由于气动肌肉压力-位移关系的非线性和时变性,研究者们很难从理论上建立一个在整个工作范围非常准确,而且适用于各种类型气动肌肉的数学模型。本文从具体的使用条件出发,建立了一种实用、简单、有效的气动肌肉驱动特性模型。
单根气动肌肉只能提供拉力,而且由于编织套和内部橡胶气腔的摩擦损耗,单根气动肌肉驱动存在很大的回滞现象。因此在实际应用中,多采用气动肌肉对拉的形式驱动机器人关节。这种对拉的驱动方式借鉴的是生物骨骼肌的原理。本文以康复机器人的肘关节为模型,分析研究了肘关节刚度和位置关于气动肌肉内部气压之间的关系,提出了一种位置与刚度混合控制的策略。
对气动肌肉分析研究后,本文开始着手搭建适用于康复训练机器人的嵌入式控制系统。首先分析了该控制器的功能需求,并以此为基础完成了中央处理器的选型。采用CORTEX M3微控制器作为系统的主芯片是本文的特点之一。然后分模块具体讨论嵌入式控制器的各主要功能,包括ADC模块、DAC模块、数据存储模块和以太网通信模块等。AD/DA转换是本控制器的核心功能,本文也采取多种措施尽可能提高转换精度。该控制器移植了FatFs文件系统,因此能够便利的与上位机共享文件。最后编写了基于前后台结构的系统软件,并给出了主要程序的关键代码。
The pneumatic muscle actuator (PMA) is such a system having performance characteristics equal to or better than that of organic muscle. It has a very great potential in medical rehabilitation robots, because compliance and elasticity have to be considered in their actuators to get robot’s safety. In this paper we describe the model of the PMA, and discuss the way to control the stiffness and position of the joint that is actuated by an antagonistic muscle pair. After that we design an embedded controller with the latest evolution of ARM’s embedded cores– CORTEX M3, for a rehabilitation robot which is driven by the PMA.
However, due to its inherent nonlinearities and time lag, this actuator suffers from poor position and force control. The researchers are very difficult to establish one accurate mathematical model in the entire operating region. Therefore, it is necessity to establish a practical, simple, and effective PMA model subject to a domain-related condition.
A joint actuated by two antagonistic PMAs is common and effective, just like the principle of human arm behavior. Considering a joint that is actuated by an antagonistic muscle pair, we analyse the relationship between the stiffness, position of the joint and the pressure inside the PMAs. Later, we give a control strategy to regulate both the stiffness and position simultaneously.
After that we begin to build the rehabilitation robot controller. First we analyzed the function needed, and choose the CORTEX M3 as the central processor. Then each major module is discussed specifically, including the ADC module, DAC module, data storage module and ethernet communication module and so on. The ADC/DAC is the controller's main function, so we try all kinds of measures to increase the conversion accuracy as far as possible. A file system named FatFs is transplanted in our controller. Therefore the PC can share the file stored in the controller easily. Finally we write a control software based on the foreground-background processing, and give the key code.
由于气动肌肉压力-位移关系的非线性和时变性,研究者们很难从理论上建立一个在整个工作范围非常准确,而且适用于各种类型气动肌肉的数学模型。本文从具体的使用条件出发,建立了一种实用、简单、有效的气动肌肉驱动特性模型。
单根气动肌肉只能提供拉力,而且由于编织套和内部橡胶气腔的摩擦损耗,单根气动肌肉驱动存在很大的回滞现象。因此在实际应用中,多采用气动肌肉对拉的形式驱动机器人关节。这种对拉的驱动方式借鉴的是生物骨骼肌的原理。本文以康复机器人的肘关节为模型,分析研究了肘关节刚度和位置关于气动肌肉内部气压之间的关系,提出了一种位置与刚度混合控制的策略。
对气动肌肉分析研究后,本文开始着手搭建适用于康复训练机器人的嵌入式控制系统。首先分析了该控制器的功能需求,并以此为基础完成了中央处理器的选型。采用CORTEX M3微控制器作为系统的主芯片是本文的特点之一。然后分模块具体讨论嵌入式控制器的各主要功能,包括ADC模块、DAC模块、数据存储模块和以太网通信模块等。AD/DA转换是本控制器的核心功能,本文也采取多种措施尽可能提高转换精度。该控制器移植了FatFs文件系统,因此能够便利的与上位机共享文件。最后编写了基于前后台结构的系统软件,并给出了主要程序的关键代码。
The pneumatic muscle actuator (PMA) is such a system having performance characteristics equal to or better than that of organic muscle. It has a very great potential in medical rehabilitation robots, because compliance and elasticity have to be considered in their actuators to get robot’s safety. In this paper we describe the model of the PMA, and discuss the way to control the stiffness and position of the joint that is actuated by an antagonistic muscle pair. After that we design an embedded controller with the latest evolution of ARM’s embedded cores– CORTEX M3, for a rehabilitation robot which is driven by the PMA.
However, due to its inherent nonlinearities and time lag, this actuator suffers from poor position and force control. The researchers are very difficult to establish one accurate mathematical model in the entire operating region. Therefore, it is necessity to establish a practical, simple, and effective PMA model subject to a domain-related condition.
A joint actuated by two antagonistic PMAs is common and effective, just like the principle of human arm behavior. Considering a joint that is actuated by an antagonistic muscle pair, we analyse the relationship between the stiffness, position of the joint and the pressure inside the PMAs. Later, we give a control strategy to regulate both the stiffness and position simultaneously.
After that we begin to build the rehabilitation robot controller. First we analyzed the function needed, and choose the CORTEX M3 as the central processor. Then each major module is discussed specifically, including the ADC module, DAC module, data storage module and ethernet communication module and so on. The ADC/DAC is the controller's main function, so we try all kinds of measures to increase the conversion accuracy as far as possible. A file system named FatFs is transplanted in our controller. Therefore the PC can share the file stored in the controller easily. Finally we write a control software based on the foreground-background processing, and give the key code.
引文
[1]中村隆一.脑卒中康复.上海:华东师范大学出版社, 1994
[2]庞翔,赵丽繁.康复训练在偏瘫康复中的意义.实用医技杂志, 2006, 13(24): 4379~4380
[3]熊有伦.机器人技术基础.武汉:华中科技大学出版社, 2004
[4]吴广玉,姜复兴.机器人工程导论.哈尔滨:哈尔滨工业大学出版社, 1988
[5]殷跃红,尉忠信,朱剑英.机器人柔顺控制研究.机器人, 1998, 20(3): 232~240
[6]傅晓云,方敏,李宝仁.气动人工肌肉驱动的直线关节的研究.机床与液压, 2006, 18: 75~82
[7]隋立明,张立勋.气动技术在康复领域的应用.液压气动与密封, 2006, 4: 33~35
[8]卫玉芬.气动肌肉驱动机器人手臂的设计与控制研究:博士学位论文。南京:南京理工大学, 2006
[9] Keith E. Gordon, Gregory S. Sawicki, Daniel P. Ferris. Mechanical performance of artificial pneumatic muscles to power an ankle-foot orthosis. Journal of Biomechanics, 2006, 39: 1832–1841
[10] Deepak Trivedi, Dustin Dienno, Christopher D. Rahn. Optimal, Model-Based Design of Soft Robotic Manipulators. Journal of Mechanical Design, SEP. 2008, Vol. 130 / 091402-1
[11] Mich?el Van Damme, Bram Vanderborght, Bjorn Verrelst et al. Proxy-based Sliding Mode Control of a Planar Pneumatic Manipulator. The International Journal of Robotics Research, 2009, 28: 266
[12] Bram Vanderborght, Bjorn Verrelst, Ronald Van Ham et al. Objective locomotion parameters based inverted pendulum trajectory generator. Robotics and Autonomous Systems, 2008, 56: 738–750
[13]隋立明,王祖温,包钢.气动肌肉驱动仿人臂的设计.液压与气动, 2004, 9: 7~8
[14] Xiaocong Zhua, Guoliang Tao, Bin Yao et al. Adaptive robust posture control of a parallel manipulator driven by pneumatic muscles. Automatica, 2008, 44: 2248–2257
[15] Xuan-yin Wang, Yang Zhang, Xiao-jie Fu et al. Design and Kinematic Analysis of a Novel Humanoid Robot Eye Using Pneumatic Artificial Muscles. Journal of Bionic Engineering, 2008, 5: 264–270
[16] Steven Gerold Northup. Biologically inspired control of a humanoid robot with nonlinear actuators. Doctor thesis. Vanderbilt University, 2001
[17]林良明,田社平.颜国正.医疗机器人用空气压人工肌线性控制的研究.中国医疗器械杂志, 2002, 26(1): 7~10
[18]王祖温,隋立明,包钢.气动肌肉驱动关节的输入整形研究.机械工程学报, 2005, 41(l): 66~70
[19] Pack R. T. , Iskarous M. , and Kawamura K. . Comparison of fuzzy and nonlinear control techniques for a flexible rubbertuator-based robot joint. Proeeedings of the International Fuzzy Systems and Intelligent Control Conefrenee, Institute of Fuzzy Systems and Intelligent Control, May. , 1994: 361-370
[20] Carbonell P. , Jiang Z. P. , and Reppegrer D. W. . A fuzzy back stepping controller for a Pneumatic muscle actuator system. Proceedings of the IEEE international Symposium on Intelligent Control, Sep. , 2001: 353-358
[21] Northrup S. , Brown Jr. E. E. , and Parlaktuna O. et al. Biologically Inspired Control Architecture for an Upper Limb.Intelligent Robotic Orthosis,International Journal of Human-friendly Welfare Robotic Systems, 2001, 2(3): 4-8
[22]陆雯.绿色化气动执行元件一气动肌腱及其应用研:硕士学位论文。苏州:苏州大学, 2006
[23] Hannaford, Glenn K, and Klute et al.Fatigue Characteristic of McKibben Artificial Muscle Actuators. Proceedings of the 1998 IEEE/RSJ International Confrence on Intelligent Robots and Systems, Victoria B. C. Canada. 1998
[24] Festo Fluidic Muscle使用手册
[25] Chou Ching-Ping, Hannaford B. Measurement and Modeling of McKibben Pnernatic artificial Muscles. IEEE Transactions on Robotics and Automation, 1996, 2, vo1. 12-1: 90-102
[26] Tsagarakis N. , Caldwell.Improved modeling and assessment of pneumatic musele actuators. Proeeedings of the IEEE international conference on robotics and automation, 2000: 3641-3646
[27] Klute G. K. , Hannaford B.Accounting for elastic energy storage in mckibben artificial muscle actuators. ASME Journal of Dynamic Systems Measurement and Control, Jun. , 2000: 386-388
[28] Bertrand Tondu, Pierre Lopez. Modeling and control of Mckibben artificial musele robot actuators. IEEE Control Systems Magazine, 2000, 20(2): 15-38
[29] D. B. REYNOLDS, D. W. REPPERGER, and C. A. PHILLIPS et al. Modeling theDynamic Characteristics of Pneumatic Muscle. Annals of Biomedical Engineering, 2003, Vol. 31: 310–317
[30]李醒飞,郑奇,张国雄.类人气动肌肉模型与实验研究.天津大学学报, 2005, 38(3): 242-247
[31] Harald Aschemann, Dominik Schindele. Sliding-Mode Control of a High-Speed Linear Axis Driven by Pneumatic Muscle Actuators. IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS, NOV. , 2008, 55(11): 3855-3864
[32]胡宇川,季林红,从医学角度探讨偏瘫上肢康复训练机器人的设计.中国临床康复, 2004. 8(37): 7754-7756
[33]隋立明,王祖温,包钢.气动肌肉的刚度特性分析.中国机械工程, 2004, 3(15): 242-244
[34]杨明,肘关节康复训练装置的研究:硕士学位论文。武汉:华中科技大学, 2008
[35] D. W. Repperger, C. A. Phillips, and A. Neidhard-Doll et al. Actuator design using biomimicry methods and a pneumatic muscle system. Control Engineering Practice, 2006, 14: 999–1009
[36]丁伯坦.残疾患者的康复治疗和训练.医疗保健器具, 1997, 6: 6-7
[37] Chad English, Donald Russell. Implementation of variable joint stiffness through antagonistic actuation using rolamite springs. Mechanism and Machine Theory, 1999, 24: 27-40
[38]昊迪,郝军,沙毅等.嵌入式系统原理设计与应用.北京:机械工业出版社, 2004
[39]马忠梅. ARM嵌入式处理器结构与应用基础.北京:北京航空航天大学出版社, 2002
[40]田泽编著.嵌入式系统开发与应用.北京:北京航空航天大学出版社, 2005
[41]闫迎春.嵌入式操作系统的研究与应用:硕士学位论文。哈尔滨:哈尔滨工业大学, 2005
[42]孙天泽.基于ARM处理器的嵌入式开发平台研究与设计:硕士学位论文。长春:吉林大学, 2003
[43] Shyam Sadasivan. An Introduction to the ARM Cortex-M3 Processor. Oct. , 2006
[44] ST抓住技术先机重塑MCU市场.电子技术应用, 2007, 8: 4
[45] STM32F103xx advanced ARM-based 32-bit MCUs Reference manual
[46]陈文凯,张根宝,张震强.基于ARM Cortex--M3内核微控制器的智能型重锤式料位检测系统.计算机测量与控制, 2008, 16(11): 1564-1566
[47]问治国.基于ARM-DSP架构的医检仪器控制系统开发:硕士学位论文。南京:东南大学, 2008
[48] http://elm-chan.org/fsw/ff/00index_e.html
[2]庞翔,赵丽繁.康复训练在偏瘫康复中的意义.实用医技杂志, 2006, 13(24): 4379~4380
[3]熊有伦.机器人技术基础.武汉:华中科技大学出版社, 2004
[4]吴广玉,姜复兴.机器人工程导论.哈尔滨:哈尔滨工业大学出版社, 1988
[5]殷跃红,尉忠信,朱剑英.机器人柔顺控制研究.机器人, 1998, 20(3): 232~240
[6]傅晓云,方敏,李宝仁.气动人工肌肉驱动的直线关节的研究.机床与液压, 2006, 18: 75~82
[7]隋立明,张立勋.气动技术在康复领域的应用.液压气动与密封, 2006, 4: 33~35
[8]卫玉芬.气动肌肉驱动机器人手臂的设计与控制研究:博士学位论文。南京:南京理工大学, 2006
[9] Keith E. Gordon, Gregory S. Sawicki, Daniel P. Ferris. Mechanical performance of artificial pneumatic muscles to power an ankle-foot orthosis. Journal of Biomechanics, 2006, 39: 1832–1841
[10] Deepak Trivedi, Dustin Dienno, Christopher D. Rahn. Optimal, Model-Based Design of Soft Robotic Manipulators. Journal of Mechanical Design, SEP. 2008, Vol. 130 / 091402-1
[11] Mich?el Van Damme, Bram Vanderborght, Bjorn Verrelst et al. Proxy-based Sliding Mode Control of a Planar Pneumatic Manipulator. The International Journal of Robotics Research, 2009, 28: 266
[12] Bram Vanderborght, Bjorn Verrelst, Ronald Van Ham et al. Objective locomotion parameters based inverted pendulum trajectory generator. Robotics and Autonomous Systems, 2008, 56: 738–750
[13]隋立明,王祖温,包钢.气动肌肉驱动仿人臂的设计.液压与气动, 2004, 9: 7~8
[14] Xiaocong Zhua, Guoliang Tao, Bin Yao et al. Adaptive robust posture control of a parallel manipulator driven by pneumatic muscles. Automatica, 2008, 44: 2248–2257
[15] Xuan-yin Wang, Yang Zhang, Xiao-jie Fu et al. Design and Kinematic Analysis of a Novel Humanoid Robot Eye Using Pneumatic Artificial Muscles. Journal of Bionic Engineering, 2008, 5: 264–270
[16] Steven Gerold Northup. Biologically inspired control of a humanoid robot with nonlinear actuators. Doctor thesis. Vanderbilt University, 2001
[17]林良明,田社平.颜国正.医疗机器人用空气压人工肌线性控制的研究.中国医疗器械杂志, 2002, 26(1): 7~10
[18]王祖温,隋立明,包钢.气动肌肉驱动关节的输入整形研究.机械工程学报, 2005, 41(l): 66~70
[19] Pack R. T. , Iskarous M. , and Kawamura K. . Comparison of fuzzy and nonlinear control techniques for a flexible rubbertuator-based robot joint. Proeeedings of the International Fuzzy Systems and Intelligent Control Conefrenee, Institute of Fuzzy Systems and Intelligent Control, May. , 1994: 361-370
[20] Carbonell P. , Jiang Z. P. , and Reppegrer D. W. . A fuzzy back stepping controller for a Pneumatic muscle actuator system. Proceedings of the IEEE international Symposium on Intelligent Control, Sep. , 2001: 353-358
[21] Northrup S. , Brown Jr. E. E. , and Parlaktuna O. et al. Biologically Inspired Control Architecture for an Upper Limb.Intelligent Robotic Orthosis,International Journal of Human-friendly Welfare Robotic Systems, 2001, 2(3): 4-8
[22]陆雯.绿色化气动执行元件一气动肌腱及其应用研:硕士学位论文。苏州:苏州大学, 2006
[23] Hannaford, Glenn K, and Klute et al.Fatigue Characteristic of McKibben Artificial Muscle Actuators. Proceedings of the 1998 IEEE/RSJ International Confrence on Intelligent Robots and Systems, Victoria B. C. Canada. 1998
[24] Festo Fluidic Muscle使用手册
[25] Chou Ching-Ping, Hannaford B. Measurement and Modeling of McKibben Pnernatic artificial Muscles. IEEE Transactions on Robotics and Automation, 1996, 2, vo1. 12-1: 90-102
[26] Tsagarakis N. , Caldwell.Improved modeling and assessment of pneumatic musele actuators. Proeeedings of the IEEE international conference on robotics and automation, 2000: 3641-3646
[27] Klute G. K. , Hannaford B.Accounting for elastic energy storage in mckibben artificial muscle actuators. ASME Journal of Dynamic Systems Measurement and Control, Jun. , 2000: 386-388
[28] Bertrand Tondu, Pierre Lopez. Modeling and control of Mckibben artificial musele robot actuators. IEEE Control Systems Magazine, 2000, 20(2): 15-38
[29] D. B. REYNOLDS, D. W. REPPERGER, and C. A. PHILLIPS et al. Modeling theDynamic Characteristics of Pneumatic Muscle. Annals of Biomedical Engineering, 2003, Vol. 31: 310–317
[30]李醒飞,郑奇,张国雄.类人气动肌肉模型与实验研究.天津大学学报, 2005, 38(3): 242-247
[31] Harald Aschemann, Dominik Schindele. Sliding-Mode Control of a High-Speed Linear Axis Driven by Pneumatic Muscle Actuators. IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS, NOV. , 2008, 55(11): 3855-3864
[32]胡宇川,季林红,从医学角度探讨偏瘫上肢康复训练机器人的设计.中国临床康复, 2004. 8(37): 7754-7756
[33]隋立明,王祖温,包钢.气动肌肉的刚度特性分析.中国机械工程, 2004, 3(15): 242-244
[34]杨明,肘关节康复训练装置的研究:硕士学位论文。武汉:华中科技大学, 2008
[35] D. W. Repperger, C. A. Phillips, and A. Neidhard-Doll et al. Actuator design using biomimicry methods and a pneumatic muscle system. Control Engineering Practice, 2006, 14: 999–1009
[36]丁伯坦.残疾患者的康复治疗和训练.医疗保健器具, 1997, 6: 6-7
[37] Chad English, Donald Russell. Implementation of variable joint stiffness through antagonistic actuation using rolamite springs. Mechanism and Machine Theory, 1999, 24: 27-40
[38]昊迪,郝军,沙毅等.嵌入式系统原理设计与应用.北京:机械工业出版社, 2004
[39]马忠梅. ARM嵌入式处理器结构与应用基础.北京:北京航空航天大学出版社, 2002
[40]田泽编著.嵌入式系统开发与应用.北京:北京航空航天大学出版社, 2005
[41]闫迎春.嵌入式操作系统的研究与应用:硕士学位论文。哈尔滨:哈尔滨工业大学, 2005
[42]孙天泽.基于ARM处理器的嵌入式开发平台研究与设计:硕士学位论文。长春:吉林大学, 2003
[43] Shyam Sadasivan. An Introduction to the ARM Cortex-M3 Processor. Oct. , 2006
[44] ST抓住技术先机重塑MCU市场.电子技术应用, 2007, 8: 4
[45] STM32F103xx advanced ARM-based 32-bit MCUs Reference manual
[46]陈文凯,张根宝,张震强.基于ARM Cortex--M3内核微控制器的智能型重锤式料位检测系统.计算机测量与控制, 2008, 16(11): 1564-1566
[47]问治国.基于ARM-DSP架构的医检仪器控制系统开发:硕士学位论文。南京:东南大学, 2008
[48] http://elm-chan.org/fsw/ff/00index_e.html