行走辅助训练机器人研究
|
摘要
行走辅助训练机器人是一种能够帮助具有运动功能障碍患者和老年人进行辅助行走的康复训练设备，在控制下肢髋关节、膝关节运动符合正常步态的同时，控制骨盆的运动规律，并与下肢步态相协调，从而实现训练者自身平衡功能，达到更好的助行训练目标。研究行走辅助训练机器人对开展康复训练技术、提高患者和老年人的生活质量具有重要意义，也符合国家中长期科学和技术发展规划纲要要求。
本文在“863”重点课题“多功能助行康复机器人”项目资助下，针对行走辅助训练机器人的结构、系统机构学特性和控制策略进行了理论分析和仿真研究，并研制了行走辅助训练机器人实验样机，基于dSPACE半物理仿真平台对机器人进行了实验研究。
在综合分析了国内外有关下肢康复训练机器人和骨盆运动控制机器人技术的基础上，提出了一种行走辅助训练机器人，该机器人包括串并联欠驱动骨盆运动控制机构和绳索牵引下肢运动控制机构，能够实现下肢髋关节、膝关节运动控制和骨盆运动控制，且允许训练者的主动性运动，使机器人更好地满足训练安全性要求。
针对骨盆运动控制机构，通过建立完全驱动情况下的系统运动学模型，规划操作端运动轨迹，并对欠驱动情况下的控制输出进行精度分析。在此基础上分析了机构惯性对驱动特性的影响，规划了期望轨迹下的驱动力矩，仿真研究了对中约束弹簧刚度和预紧力对骨盆运动特性的影响。针对下肢运动控制机构，利用影响系数法分析了绳索运行速度、加速度与髋关节、膝关节运行速度、加速度之间的映射关系，并通过机构仿真手段规划了期望轨迹的绳索运动输出；利用约束法建立了系统力螺旋平衡方程，基于达朗伯原理分析了系统动力学特性，通过广义逆求得绳索拉力解，为驱动电机选择和驱动特性实验提供依据。
考虑行走辅助训练机器人是多输入多输出非线性系统，且具有强耦合的动力学特性。在分析鲁棒跟踪控制策略特点基础上，提出鲁棒跟踪控制策略实现对行走辅助训练机器人的跟踪控制。由于行走辅助训练机器人下肢运动运动控制系统模型的建模的不准确和受到外界干扰，将行走辅助训练机器人控制系统考虑为一类不确定性线性系统，将系统的不确定性假设满足广义的匹配条件，在此基础上设计鲁棒跟踪控制器。通过对下肢运动控制机构的轨迹鲁棒跟踪控制仿真，分析了系统控制精度，证明了鲁棒跟踪控制策略的可行性。
根据行走辅助训练机器人结构方案，研制了行走辅助训练机器人实验样机。以dSPACE半物理仿真平台作为控制系统的上位机，对骨盆运动控制机构进行了电动缸实验、系统标定实验，在此基础上对系统控制实验进行了详细的分析；对下肢运动控制机构在确定控制器参数基础上，进行了各种单项实验和步态轨迹跟踪实验。实验结果表明行走辅助训练机器人整体性能满足实际应用要求。
Walking aid training robot is a kind of rehabilitation training equipment, which can assistpatients with motor dysfunction and elderly people to walk. It can control the motion of pelvisand coordinate with lower limbs’ gait while making hip or knee motion coordinate withnormal gait, thereby the self-balance of trainee can be acheived and the training goal can berealized. The study on walking aid training robot is significant to develop rehabilitationtraining technology and improve the life quality of patient and elderly people, meanwhile, it isalso in accordance with the requirements of the National long-term plan for science andtechnology development.
Supported by the project “multi-functional walking aid rehabilitation robot”, which isone of the863key projects, this dissertation theoretically analyzes and simulates the structureof robot, the features of systematic mechanism and controlling strategies, manufactures theprototype, and experimentally studies the robot based on dSPACE semi-physical simulationplatform.
A kind of walking aid training robot is put forward based on the review and analysis oftechnologies about lower limb rehabilitation training robot and pelvis motion control robotboth home and abroad. This robot includes series-parallel connection underactuated pelvismotion control mechanism and rope pull lower limb motion control mechanism, which canrealize the motion control of lower limb hip or knee and pelvis, and allow initiative movementof trainee that better fulfills the security requirements of training.
For pelvis motion control mechanism a completely driven kinematics model isestablished, the operation trajectory planned and the accuracy of underactuated control outputanalyzed. Then the effect of mechanism inertia on drive performance is discussed, the drivingmoment of expected track planned and the effect of centring constraint spring stiffness andpretightening force on pelvis motion simulated. For lower limb control mechanism themapping relationship between rope motion speed/acceleration and hip or knee motionspeed/acceleration is analyzed by influence coefficient method, and rope motion output ofexpected track also planned by mechanism simulation. By using restraint method systematicforce spiral balance equation is established. Based on d'alembert's principle system dynamicperformance is analyzed. And rope tension solution is acquired by generalized inverse algorithm. All the above provides basis for the choice of drive motor and the driveperformance test.
Considering walking aid training robot is a multiple input multiple output nonlinearsystem, and has a strong coupling dynamic characteristics. In the analysis of robust trackingcontrol strategy is proposed based on the characteristics, a robust tracking control strategy toachieve walking aid training robot tracking control. Because walking aid training robot lowerlimb motor motion control system modeling is not accurate and external disturbances, thewalking aid training robot control system is considered for a class of uncertain linear systems,the system uncertainty hypothesis satisfies generalized matching condition, based on thedesign of robust tracking controller. The lower limb motion control mechanism track robusttracking control simulation, analysis of the accuracy of the control system, proved thefeasibility of robust tracking control strategy.
According to the structural concept of walking aid training robot, the prototype ismanufactured. The upper computer of control system is dSPACE semi-physical simulationplatform, with which electric cylinder test and system calibration test are carried out on pelvismotion control mechanism, and system control test is analyzed in detail. A variety ofindividual tests and gait trajectory tracking test are carried out on lower limb motion controlmechanism after the confirmation of controller parameters. The test results indicate that theoverall performance of walking aid training robot can satisfy the requirements of practicaluse.
本文在“863”重点课题“多功能助行康复机器人”项目资助下，针对行走辅助训练机器人的结构、系统机构学特性和控制策略进行了理论分析和仿真研究，并研制了行走辅助训练机器人实验样机，基于dSPACE半物理仿真平台对机器人进行了实验研究。
在综合分析了国内外有关下肢康复训练机器人和骨盆运动控制机器人技术的基础上，提出了一种行走辅助训练机器人，该机器人包括串并联欠驱动骨盆运动控制机构和绳索牵引下肢运动控制机构，能够实现下肢髋关节、膝关节运动控制和骨盆运动控制，且允许训练者的主动性运动，使机器人更好地满足训练安全性要求。
针对骨盆运动控制机构，通过建立完全驱动情况下的系统运动学模型，规划操作端运动轨迹，并对欠驱动情况下的控制输出进行精度分析。在此基础上分析了机构惯性对驱动特性的影响，规划了期望轨迹下的驱动力矩，仿真研究了对中约束弹簧刚度和预紧力对骨盆运动特性的影响。针对下肢运动控制机构，利用影响系数法分析了绳索运行速度、加速度与髋关节、膝关节运行速度、加速度之间的映射关系，并通过机构仿真手段规划了期望轨迹的绳索运动输出；利用约束法建立了系统力螺旋平衡方程，基于达朗伯原理分析了系统动力学特性，通过广义逆求得绳索拉力解，为驱动电机选择和驱动特性实验提供依据。
考虑行走辅助训练机器人是多输入多输出非线性系统，且具有强耦合的动力学特性。在分析鲁棒跟踪控制策略特点基础上，提出鲁棒跟踪控制策略实现对行走辅助训练机器人的跟踪控制。由于行走辅助训练机器人下肢运动运动控制系统模型的建模的不准确和受到外界干扰，将行走辅助训练机器人控制系统考虑为一类不确定性线性系统，将系统的不确定性假设满足广义的匹配条件，在此基础上设计鲁棒跟踪控制器。通过对下肢运动控制机构的轨迹鲁棒跟踪控制仿真，分析了系统控制精度，证明了鲁棒跟踪控制策略的可行性。
根据行走辅助训练机器人结构方案，研制了行走辅助训练机器人实验样机。以dSPACE半物理仿真平台作为控制系统的上位机，对骨盆运动控制机构进行了电动缸实验、系统标定实验，在此基础上对系统控制实验进行了详细的分析；对下肢运动控制机构在确定控制器参数基础上，进行了各种单项实验和步态轨迹跟踪实验。实验结果表明行走辅助训练机器人整体性能满足实际应用要求。
Walking aid training robot is a kind of rehabilitation training equipment, which can assistpatients with motor dysfunction and elderly people to walk. It can control the motion of pelvisand coordinate with lower limbs’ gait while making hip or knee motion coordinate withnormal gait, thereby the self-balance of trainee can be acheived and the training goal can berealized. The study on walking aid training robot is significant to develop rehabilitationtraining technology and improve the life quality of patient and elderly people, meanwhile, it isalso in accordance with the requirements of the National long-term plan for science andtechnology development.
Supported by the project “multi-functional walking aid rehabilitation robot”, which isone of the863key projects, this dissertation theoretically analyzes and simulates the structureof robot, the features of systematic mechanism and controlling strategies, manufactures theprototype, and experimentally studies the robot based on dSPACE semi-physical simulationplatform.
A kind of walking aid training robot is put forward based on the review and analysis oftechnologies about lower limb rehabilitation training robot and pelvis motion control robotboth home and abroad. This robot includes series-parallel connection underactuated pelvismotion control mechanism and rope pull lower limb motion control mechanism, which canrealize the motion control of lower limb hip or knee and pelvis, and allow initiative movementof trainee that better fulfills the security requirements of training.
For pelvis motion control mechanism a completely driven kinematics model isestablished, the operation trajectory planned and the accuracy of underactuated control outputanalyzed. Then the effect of mechanism inertia on drive performance is discussed, the drivingmoment of expected track planned and the effect of centring constraint spring stiffness andpretightening force on pelvis motion simulated. For lower limb control mechanism themapping relationship between rope motion speed/acceleration and hip or knee motionspeed/acceleration is analyzed by influence coefficient method, and rope motion output ofexpected track also planned by mechanism simulation. By using restraint method systematicforce spiral balance equation is established. Based on d'alembert's principle system dynamicperformance is analyzed. And rope tension solution is acquired by generalized inverse algorithm. All the above provides basis for the choice of drive motor and the driveperformance test.
Considering walking aid training robot is a multiple input multiple output nonlinearsystem, and has a strong coupling dynamic characteristics. In the analysis of robust trackingcontrol strategy is proposed based on the characteristics, a robust tracking control strategy toachieve walking aid training robot tracking control. Because walking aid training robot lowerlimb motor motion control system modeling is not accurate and external disturbances, thewalking aid training robot control system is considered for a class of uncertain linear systems,the system uncertainty hypothesis satisfies generalized matching condition, based on thedesign of robust tracking controller. The lower limb motion control mechanism track robusttracking control simulation, analysis of the accuracy of the control system, proved thefeasibility of robust tracking control strategy.
According to the structural concept of walking aid training robot, the prototype ismanufactured. The upper computer of control system is dSPACE semi-physical simulationplatform, with which electric cylinder test and system calibration test are carried out on pelvismotion control mechanism, and system control test is analyzed in detail. A variety ofindividual tests and gait trajectory tracking test are carried out on lower limb motion controlmechanism after the confirmation of controller parameters. The test results indicate that theoverall performance of walking aid training robot can satisfy the requirements of practicaluse.
引文
[1]袁启明.蓬勃发展的康复医学工程.医疗保健器具.1999,(6):31-33页
[2] http://www.chelder.com.cn/zhanhui/2010/69498.html
[3] http://cn.chinagate.cn/population/2009-10/26/content_18767047.htm
[4] http://www.shouxinggong.com/fuwu/fuwu-20100329-4673.html
[5] http://www.chinapop.gov.cn/xwzx/rkxw/200910/t20091026_182234.html
[6]金德文，张济川.康复工程学的研究与发展.现代康复.2000,4(5):643-646页
[7] http://www.most.gov.cn/tztg/t20061001_36446.htm
[8] http://www.carm.org.cn/index.do#
[9]王克义.绳索牵引骨盆运动控制康复机器人研究.哈尔滨工程大学.2009:32-38页
[10]韩雯，常均.现代康复工程的发展概述.医疗装备.2007,(4):13-15页
[11]杜志江，孙传杰，陈艳宁.康复机器人研究现状.中国康复医学杂志.2003,18(5):293-284页
[12] A M Simon, G R Brent, D P Ferris. Symmetry-based resistance as a novel means oflower limb rehabilitation. Journal of Biomechanics.2007,40(6):1286-1292P
[13] V Dietz, R Mueller, G Colombo. Locomotor activity in spinal man: Significance ofafferent input from joint and load receptors.Brain.2002,125(12):2626-2634P
[14] AC Lo, E W Triche. Improving gait in multiple sclerosis using robot-assisted, bodyweight supported treadmill training.Neurorehabil Neural Repair.2008,22(6):661-671P
[15] L Lünenburger, G Colombo, R Riener, et al. Clinical assessments performed duringrobotic rehabilitation by the gait training robot Lokomat. Proceedings of the2005IEEE9th International Conference on Rehabilitation Robotics. Chicago, USA,2005:345-348P
[16] H Schmidt, D Sorowka, S Hesse, et al. Development of a robotic walking sumulator forgait rehabilitation, Biomed Tech.2003,48(10):281-286P
[17] H Schmid, S Hesse, R Bernhardt, et al. HapticWalker—A Novel Haptic Foot DeviceACM Transactions on Applied Perception.2005,2(2):166-180P
[18] R Riener, T Nef, G Colombo. Robot-aided neuro rehabilitation of the upper extremities.Med Biol Eng Comput.2005,43(1):2-10P
[19] S Hesse, G T Schulte, M Konrad,et al.Robot-assisted arm trainer for the passive andactive practice of bilateral forearm and wrist movements in hemiparetic subjects. ArchPhys Med Rehabil.2003,84(6):915-920P.
[20] Y Stauffer, F Reynard, Y Allemand, et al. Pelvic motion implementation on theWalkTrainer.2007IEEE International Conference on Robotics and Biomimetics.2007:133-138P
[21] H Schmidt, F Piorko, R Bernhardt, et al. Synthesis of perturbations for gait rehabilitationrobots. Proceedings of the2005IEEE9th International Conference on RehabilitationRobotics. Chicago,2005:74-77P
[22] K Homma, O Fukuda, Y Nagata, et al. Study of a wire-driven leg rehabilitation system.2004IEEE/RSJ International Conference on Intelligent Robots and Systems,2004:1668-1673P
[23] R Verhoeven.Analysis of the Workspace of Tendon-based Stewart Platforms. Duisburg:Gerhard MercatorUniversity.2004:27-28P
[24] H Yamaura, K Matsushita, R Kato, et al. Development of Hand Rehabilitation SystemUsing Wire-Driven Link Mechanism for Paralysis Patients. Proceedings of the2009IEEE International Conference on Robotics and Biominetics. China: Guilin,2009:19-23P
[25] J L Emken, J H Wynne, S J Harkema, et al. A robotic device for manipulating humanstepping. IEEE Transactions on Robotics.2006,22(1):185-189P.
[26] D J Reinkensmeyer, D Aoyagi, J L Emken, et al. Tools for understanding and optimizingrobotic gait training. Journal of Rehabilitation Research and Development.2006,43(5):657-670P
[27] J M Baydal. Development of a new concept of an orthotic knee joint compatible with thekinematics of the human knee joint. Prosthetic and Orthotic International.2006:371-383P
[28] B Preising, T C Hsia, B Mittelstadt. A literaturere view: robots in medicine.IEEEEngineering in Medicineand Biology.1991,10(2):13-22P
[29] J F Veneman, R Kruidhof, E E G Hekman, et al. Design and Evaluation of the LOPESExoskeleton Robot for Interactive Gait Rehabilitation. IEEE transactions on neuralsystems and rehabilitation engineering.2007,15(3):379-386P
[30] R Ekkelenkamp, J Veneman, H V D Kooij. LOPES: selective control of gait functionsduring the gait rehabilitation of CVA patients. Rehabilitation Robotics ICORR,9thInternational Conference,2005:361-364P
[31] J F Veneman, R Kruidhof, E E G Hekman, et al. Design and Evaluation of the LOPESExoskeleton Robot for Interactive Gait Rehabilitation. IEEE Transactions on neuralsystems and rehabilitation engineering.2007,15(3):379-386P
[32] R kkelenkamp, J Veneman, H V D Kooij. LOPES: A lower extremity poweredexoskeleton,2007IEEE International Conference on Robotics and Automation, Rome.2007:3132-3133P
[33] H Kobayshi. Takeo Karato Toshiaki Tsuji. International Conference on Mechatronics andAutomation.2007:187-191P
[34] K Kong，M Tomizuka. Flexible Joint Actuator for Patient’s Rehabilitation Device.16thIEEE International Conference on Robot and Human Interactive Communication. Korea:Jeju,2007:1179-1184P
[35] Dariop, Guglielmellie, Laschic. Humanoids and personal robot: Design and experiments.Journal of Robotic System.2001,8(12)673-690P
[36]丁敏，李建民，吴庆文等.下肢步态康复机器人：研究进展及临床应用.中国组织工程研究与临床康复.2010,14(35):6604-6607页
[37]李震，于学龙.新型下肢关节康复器的研制.中国医疗器械信息.2006,12(12):35-38页
[38]马俊宝.下肢康复机器人放摔倒控制实验研究.沈阳工业大学硕士论文.2011:14-16页
[39]王喜太，陈云飞，王责兰等.髓关节康复训练器控制系统设计与研究.现代科学仪器.2009,5(10):23-25页
[40]杨俊友，白殿春，王硕玉等.全方位轮式下肢康复训练机器人轨迹跟踪控制.机器人.2011,33(10):314-318页
[41] http://61.153.34.35:8002/kjqk/jqrjsyyy/jqrj2004/040403.pdf
[42] http://news.tsinghua.edu.cn/new/readnews.php?id=22485
[43]王人成，王爱明，贾晓红等.一种减重步行训练康复机器人.200910082601.发明专利.2009
[44]钱晋武，冯治国.下肢康复自动步态矫正器.2007101716964.发明专利.2007
[45]孙建东，金德文.一种下肢被动运动康复训练器.中国康复医学杂志.2001,16(5):298-299页
[46]薛渊，吕广明.下肢康复助行机构本体设计及运动学分析.机械设计与制造.2006(5):131-133页
[47]薛渊，吕广明.下肢康复助行机构及其液压系统设计.液压与气动.2006(1):45-47页
[48]胡伟佳.被动式步态康复训练器的设计与研究.哈尔滨工程大学硕士论文.2007:22-28页
[49]谭世江.气动助力机械腿机构设计及实验研究.哈尔滨工程大学硕士论文.2008:12-15页
[50]王志超.卧式下肢康复机器人的虚拟现实研究.哈尔滨工程大学硕士论文.2008:30-40页
[51]张立勋，王令军，王凤良等.步态训练机器人人机系统动力学仿真.高技术通讯.2009,19(11):1153-1158页
[52] http://www.bdhkf.com/
[53] Y Stauffer, Y Allemand, M Bouri, et al. Pelvic motion measurement during over groundwalking, analysis and implementation on the WalkTrainer reeducation device. IEEE/RSJInternational Conference on Intelligent Robots and Systems.2008:2362-2367P
[54] Y Stauffer, F Reynard, Y Allemand, et al. Pelvic motion implementation on theWalkTrainer.2007IEEE International Conference on Robotics and Biomimetics.2007:133-138P
[55] H Schmidt, C Werner, R Bernhardt, et al. Gait rehabilitation machines based onprogrammable footplates. Journal of Neuro Engineering and Rehabilitation.2007,4(2):l-7P
[56] D Surdilovic, J Zhang, R Bernhardt. STRING-MAN: Wire-robot technology for safe,flexible and human-friendly gait rehabilitation.2007IEEE10th International Conferenceon Rehabilitation Robotics. Netherlands: Noordwijk,2007:446-453P
[57] D Aoyagi, W E Ichinose, S J Harkema. An assistive robotic device that can synchronizeto the pelvic motion during human gait training. Proceedings of the2005IEEE9thInternational Conference on Rehabilitation Robotics.2005:565-568P
[58] P Michael, B A David, J S M Julio, et al. KineAssist: A robotic overground gait andbalance training device. Proceedings of the2005IEEE9th International Conference onRehabilitation Robotics,2005:241-246P
[59] H B Lim, K H Hoon, K H Low. Pelvic control and over-ground walking methodology forimpaired gait recovery. IEEE International Conference on Robotics and Biomimetics.2009:282-287P
[60] H B Lim, K H Hoon, Y C Soh. Gait planning for effective rehabilitation-from gait studyto application in clinical rehabilitation. IEEE International Conference on RehabilitationRobotics: Reaching Users&the Community.2009:271-276P
[61]赵凌燕.人体步态模式实验研究.哈尔滨工程大学博士学位论文.2008:25页
[62]张立勋，赵凌燕，王岚等.一种测量人行走时骨盆运动轨迹的新方法.哈尔滨工程大学学报.2006,27(1):128-130,135页
[63]钟巧艺，苏巴丽，陈坤嫦，等.骨盆骨折病人的护理问题及对策.中国医药指南.2008,6(2):10-12页
[64]刘文权，吴婉霞，陈坚.骨盆控制训练对脑卒中患者平衡功能和步行能力的影响.广州医药.2006,37(2):71-72页
[65]张立勋，王克义，张今瑜.人体步态周期中骨盆自由度的分析.机械设计.2008(1):12-14页
[66]王克义，张立勋，张今瑜等.1R2T绳索牵引骨盆康复机器人动力学.江苏大学学报：自然科学版.2010,31(2):131-135页
[67]王克义，张立勋，孟浩.1R2T绳索牵引并联康复机器人绳索弹性研究.南京理工大学学报:自然科学版.2010,34(5):602-607页
[68]张立勋，王克义，张今瑜等.基于绳索牵引的骨盆运动并联康复机器人的可控性研究.哈尔滨工程大学学报.2007,28(7):790-794页
[69]刘攀，张立勋，王克义等.绳牵引康复机器人的动力学分析与控制.中国机械工程.2009,20(11):1335-1339页
[70]张立勋，刘攀，王克义等.三自由度绳牵引平面并联机器人力/位控制研究.机械设计与研究.2008,24(3):22-25页
[71]张立勋，刘攀，王克义等.绳索牵引并联康复机器人的建模与控制.哈尔滨工程大学学报.2009,30(1):81-85页
[72]张琦，李琪，纪树荣.骨盆训练对脑卒中偏瘫患者步行能力的影响.中国康复理论与实践.2001,7(4):161-162页
[73]陈东辉，佟金，李重焕等.人和动物的步态与步行机器人.吉林大学学报.2003,33(4):121-125页
[74]藏红.人体运动学.人民卫生出版社,2008:20-26页
[75] D A Winter, C D Mackinnon, G K Ruder, et al. An integrated EMG biomechanical modelof upper body balance and posture during human gait. Progress in Brain Research.Amsterdam:Elsevier.1993,(97)113-156P
[76] C D Mackinnon, D A Winter. Control of whole body balance in the frontal plane duringhuman walking. Journal of Biomechanics.1993,26(6):33-44P
[77] ZHAO Lingyan, ZHANG Lixun, WANG Lan, et al. Three-dimensional motion of thepelvis during human walking. In:IEEE ICMA2005. Niagara Falls.2005:335-339P
[78] Y Stauffer. Pelvic Motion Implementation on the WalkTrainer.2007IEEE InternationalConference on Robotics and Biomimetics. Sanya,2007:133-138P
[79]古恩鹏，刘爱峰，金鸿宾等.步态分析在临床骨科与康复中的应用.中国中西医结合外科杂志.2011,17(3):335-337页
[80]王桂茂，严隽陶，刘玉超等.脑卒中偏瘫步态的时空参数与骨盆运动学分析.临床研究.2010,25(12):1148-1151页
[81] F M L Amirouche, S K Ider, J Trimble. General Analytical method for the analysis andsimulation of human locomotion. American Society of Mechanical Engineers,Bioengineering Division (Publication) BED.1988:57-60P
[82]方海峰.人体下肢有限元动力学分析模型的建立和验证.湖南大学硕士论文.2004:22页
[83] http://www.entdr.com.cn/Get/jichushuji02/163133813.htm
[84] http://www.hudong.com/wiki/%e9%ab%8b%e5%85%b3%e8%8a%82
[85]张更林，金宝士，吴雅杰.人体下肢机械式膝关节运动分析.中国临床康复.2004,8(2):210-212页
[86] W Zijlstra, A L Hof. Displacement of the pelvis during human walking: experimentaldata and model predictions. Gait and Posture.1997,(8):249-267P
[87]朱镛连.神经康复学.北京人民军医出版社,2001:154-166页
[88] T G Homby, D H Zemon, D Campbell. Robotic-assisted, body-weight-supportedtreadmill training in individuals following motor incomplete spinal cord injury. PhysicalTherapy,2005,85(1):52-66P
[89] Wang Qiyuan, Qian Jinwu, Zhang Yanan, et al. Gait Trajectory Planning and Simulationfor the Powered Gait Orthosis.2007IEEE International Conference on Robotics andBiomimetics. China: Yalong Bay, Sanya,2008:1693-1697P
[90]毕秋吉.步态训练机器人骨盆控制机构研究.哈尔滨工程大学硕士论文.2009:18-21页
[91]夏振涛.助行训练机器人系统设计及步态控制实验研究.哈尔滨工程大学硕士论文.2009:32-35页
[92]王伟.绳驱动助行机构控制.哈尔滨工程大学.2009:38-40页
[93]赵豫玉.穿戴式下肢康复机器人的研究.哈尔滨工程大学硕士论文.2009:22-24页
[94] Y Y Chew, S N Mark, N C John. Automated person recognition by walking and runningvia model-based approaches. Pattern Recognition.2004(37):1057-1072P
[95] G S Anthony, B Peter, R David. Three-dimensional angular kinematics of the lumbarspine and pelvis during running. Human Movement Science.2002(21):273-293P
[96] P H Jason, J P Anthony, J R Paul. Explicit finite element modeling of total kneereplacement mechanics. Journal of Biomechanics.2005(38):323-331P.
[97]宋伟刚.机器人学—运动学、动力学与控制.北京:科学出版社,2007:81-87页
[98]万晓正.平面绳牵引并联机构的控制研究.哈尔滨工程大学硕士论文.2009:33-36页
[99]董玉红.欠驱动力合作机器人的驱动及控制技术研究.哈尔滨工程大学博士论文.2006:42-55页
[100] J.H.金斯伯格著.白化同，李俊宝译.机械与结构振动—理论与应用.北京:中国宇航出版社,2005:12-13,22-23页
[101] Furuta, Hara, Mori. A class of systems with the same observer. IEEE Trans. Auto.Contr,1976(8):572-576P
[102] A RGralimidi, B R Barmish. The constrained Lyapunov problem and its aplication torobust output feedback stabilization. IEEE Trans.Auto.Contr.1986,31(5):410-419P
[103]梅生伟，申铁龙，刘康志.现代鲁棒控制理论与应用.北京:清华大学出版社,2003:19-22页
[104]储健，俞立，苏宏业.鲁棒控制理论及应用.杭州:浙江大学出版社,2000:55-58页
[105]俞立.鲁棒控制—线性矩阵不等式处理方法.北京:清华大学出版社,2002:18-19页
[106]俞立.鲁棒控制—线性矩阵不等式处理方法.北京:清华大学出版社,2002:27-28页
[107] L R Petersen. Quadratic stabilizability of uncertain linear systems containing bothconstant and time-varying uncertain parameters. Journal of Optimization Theory andApplications.1988,57(3):439-461P
[108] F Jabbari, W E Schmitendorf. Robust linear controllers using observers. IEE Trans AutoContr.1991,36(12):1509-1514P
[109]朱晓东，孙优贤.不确定动态时滞系统的基于观测器的鲁棒镇定设计.控制理论与应用.1996,13(2):254-257页
[110]苏宏业，王景成，褚健.一类不确定动态时滞系统的无记忆鲁棒镇定控制.自动化学报.1997,24(4):497-501页
[2] http://www.chelder.com.cn/zhanhui/2010/69498.html
[3] http://cn.chinagate.cn/population/2009-10/26/content_18767047.htm
[4] http://www.shouxinggong.com/fuwu/fuwu-20100329-4673.html
[5] http://www.chinapop.gov.cn/xwzx/rkxw/200910/t20091026_182234.html
[6]金德文，张济川.康复工程学的研究与发展.现代康复.2000,4(5):643-646页
[7] http://www.most.gov.cn/tztg/t20061001_36446.htm
[8] http://www.carm.org.cn/index.do#
[9]王克义.绳索牵引骨盆运动控制康复机器人研究.哈尔滨工程大学.2009:32-38页
[10]韩雯，常均.现代康复工程的发展概述.医疗装备.2007,(4):13-15页
[11]杜志江，孙传杰，陈艳宁.康复机器人研究现状.中国康复医学杂志.2003,18(5):293-284页
[12] A M Simon, G R Brent, D P Ferris. Symmetry-based resistance as a novel means oflower limb rehabilitation. Journal of Biomechanics.2007,40(6):1286-1292P
[13] V Dietz, R Mueller, G Colombo. Locomotor activity in spinal man: Significance ofafferent input from joint and load receptors.Brain.2002,125(12):2626-2634P
[14] AC Lo, E W Triche. Improving gait in multiple sclerosis using robot-assisted, bodyweight supported treadmill training.Neurorehabil Neural Repair.2008,22(6):661-671P
[15] L Lünenburger, G Colombo, R Riener, et al. Clinical assessments performed duringrobotic rehabilitation by the gait training robot Lokomat. Proceedings of the2005IEEE9th International Conference on Rehabilitation Robotics. Chicago, USA,2005:345-348P
[16] H Schmidt, D Sorowka, S Hesse, et al. Development of a robotic walking sumulator forgait rehabilitation, Biomed Tech.2003,48(10):281-286P
[17] H Schmid, S Hesse, R Bernhardt, et al. HapticWalker—A Novel Haptic Foot DeviceACM Transactions on Applied Perception.2005,2(2):166-180P
[18] R Riener, T Nef, G Colombo. Robot-aided neuro rehabilitation of the upper extremities.Med Biol Eng Comput.2005,43(1):2-10P
[19] S Hesse, G T Schulte, M Konrad,et al.Robot-assisted arm trainer for the passive andactive practice of bilateral forearm and wrist movements in hemiparetic subjects. ArchPhys Med Rehabil.2003,84(6):915-920P.
[20] Y Stauffer, F Reynard, Y Allemand, et al. Pelvic motion implementation on theWalkTrainer.2007IEEE International Conference on Robotics and Biomimetics.2007:133-138P
[21] H Schmidt, F Piorko, R Bernhardt, et al. Synthesis of perturbations for gait rehabilitationrobots. Proceedings of the2005IEEE9th International Conference on RehabilitationRobotics. Chicago,2005:74-77P
[22] K Homma, O Fukuda, Y Nagata, et al. Study of a wire-driven leg rehabilitation system.2004IEEE/RSJ International Conference on Intelligent Robots and Systems,2004:1668-1673P
[23] R Verhoeven.Analysis of the Workspace of Tendon-based Stewart Platforms. Duisburg:Gerhard MercatorUniversity.2004:27-28P
[24] H Yamaura, K Matsushita, R Kato, et al. Development of Hand Rehabilitation SystemUsing Wire-Driven Link Mechanism for Paralysis Patients. Proceedings of the2009IEEE International Conference on Robotics and Biominetics. China: Guilin,2009:19-23P
[25] J L Emken, J H Wynne, S J Harkema, et al. A robotic device for manipulating humanstepping. IEEE Transactions on Robotics.2006,22(1):185-189P.
[26] D J Reinkensmeyer, D Aoyagi, J L Emken, et al. Tools for understanding and optimizingrobotic gait training. Journal of Rehabilitation Research and Development.2006,43(5):657-670P
[27] J M Baydal. Development of a new concept of an orthotic knee joint compatible with thekinematics of the human knee joint. Prosthetic and Orthotic International.2006:371-383P
[28] B Preising, T C Hsia, B Mittelstadt. A literaturere view: robots in medicine.IEEEEngineering in Medicineand Biology.1991,10(2):13-22P
[29] J F Veneman, R Kruidhof, E E G Hekman, et al. Design and Evaluation of the LOPESExoskeleton Robot for Interactive Gait Rehabilitation. IEEE transactions on neuralsystems and rehabilitation engineering.2007,15(3):379-386P
[30] R Ekkelenkamp, J Veneman, H V D Kooij. LOPES: selective control of gait functionsduring the gait rehabilitation of CVA patients. Rehabilitation Robotics ICORR,9thInternational Conference,2005:361-364P
[31] J F Veneman, R Kruidhof, E E G Hekman, et al. Design and Evaluation of the LOPESExoskeleton Robot for Interactive Gait Rehabilitation. IEEE Transactions on neuralsystems and rehabilitation engineering.2007,15(3):379-386P
[32] R kkelenkamp, J Veneman, H V D Kooij. LOPES: A lower extremity poweredexoskeleton,2007IEEE International Conference on Robotics and Automation, Rome.2007:3132-3133P
[33] H Kobayshi. Takeo Karato Toshiaki Tsuji. International Conference on Mechatronics andAutomation.2007:187-191P
[34] K Kong，M Tomizuka. Flexible Joint Actuator for Patient’s Rehabilitation Device.16thIEEE International Conference on Robot and Human Interactive Communication. Korea:Jeju,2007:1179-1184P
[35] Dariop, Guglielmellie, Laschic. Humanoids and personal robot: Design and experiments.Journal of Robotic System.2001,8(12)673-690P
[36]丁敏，李建民，吴庆文等.下肢步态康复机器人：研究进展及临床应用.中国组织工程研究与临床康复.2010,14(35):6604-6607页
[37]李震，于学龙.新型下肢关节康复器的研制.中国医疗器械信息.2006,12(12):35-38页
[38]马俊宝.下肢康复机器人放摔倒控制实验研究.沈阳工业大学硕士论文.2011:14-16页
[39]王喜太，陈云飞，王责兰等.髓关节康复训练器控制系统设计与研究.现代科学仪器.2009,5(10):23-25页
[40]杨俊友，白殿春，王硕玉等.全方位轮式下肢康复训练机器人轨迹跟踪控制.机器人.2011,33(10):314-318页
[41] http://61.153.34.35:8002/kjqk/jqrjsyyy/jqrj2004/040403.pdf
[42] http://news.tsinghua.edu.cn/new/readnews.php?id=22485
[43]王人成，王爱明，贾晓红等.一种减重步行训练康复机器人.200910082601.发明专利.2009
[44]钱晋武，冯治国.下肢康复自动步态矫正器.2007101716964.发明专利.2007
[45]孙建东，金德文.一种下肢被动运动康复训练器.中国康复医学杂志.2001,16(5):298-299页
[46]薛渊，吕广明.下肢康复助行机构本体设计及运动学分析.机械设计与制造.2006(5):131-133页
[47]薛渊，吕广明.下肢康复助行机构及其液压系统设计.液压与气动.2006(1):45-47页
[48]胡伟佳.被动式步态康复训练器的设计与研究.哈尔滨工程大学硕士论文.2007:22-28页
[49]谭世江.气动助力机械腿机构设计及实验研究.哈尔滨工程大学硕士论文.2008:12-15页
[50]王志超.卧式下肢康复机器人的虚拟现实研究.哈尔滨工程大学硕士论文.2008:30-40页
[51]张立勋，王令军，王凤良等.步态训练机器人人机系统动力学仿真.高技术通讯.2009,19(11):1153-1158页
[52] http://www.bdhkf.com/
[53] Y Stauffer, Y Allemand, M Bouri, et al. Pelvic motion measurement during over groundwalking, analysis and implementation on the WalkTrainer reeducation device. IEEE/RSJInternational Conference on Intelligent Robots and Systems.2008:2362-2367P
[54] Y Stauffer, F Reynard, Y Allemand, et al. Pelvic motion implementation on theWalkTrainer.2007IEEE International Conference on Robotics and Biomimetics.2007:133-138P
[55] H Schmidt, C Werner, R Bernhardt, et al. Gait rehabilitation machines based onprogrammable footplates. Journal of Neuro Engineering and Rehabilitation.2007,4(2):l-7P
[56] D Surdilovic, J Zhang, R Bernhardt. STRING-MAN: Wire-robot technology for safe,flexible and human-friendly gait rehabilitation.2007IEEE10th International Conferenceon Rehabilitation Robotics. Netherlands: Noordwijk,2007:446-453P
[57] D Aoyagi, W E Ichinose, S J Harkema. An assistive robotic device that can synchronizeto the pelvic motion during human gait training. Proceedings of the2005IEEE9thInternational Conference on Rehabilitation Robotics.2005:565-568P
[58] P Michael, B A David, J S M Julio, et al. KineAssist: A robotic overground gait andbalance training device. Proceedings of the2005IEEE9th International Conference onRehabilitation Robotics,2005:241-246P
[59] H B Lim, K H Hoon, K H Low. Pelvic control and over-ground walking methodology forimpaired gait recovery. IEEE International Conference on Robotics and Biomimetics.2009:282-287P
[60] H B Lim, K H Hoon, Y C Soh. Gait planning for effective rehabilitation-from gait studyto application in clinical rehabilitation. IEEE International Conference on RehabilitationRobotics: Reaching Users&the Community.2009:271-276P
[61]赵凌燕.人体步态模式实验研究.哈尔滨工程大学博士学位论文.2008:25页
[62]张立勋，赵凌燕，王岚等.一种测量人行走时骨盆运动轨迹的新方法.哈尔滨工程大学学报.2006,27(1):128-130,135页
[63]钟巧艺，苏巴丽，陈坤嫦，等.骨盆骨折病人的护理问题及对策.中国医药指南.2008,6(2):10-12页
[64]刘文权，吴婉霞，陈坚.骨盆控制训练对脑卒中患者平衡功能和步行能力的影响.广州医药.2006,37(2):71-72页
[65]张立勋，王克义，张今瑜.人体步态周期中骨盆自由度的分析.机械设计.2008(1):12-14页
[66]王克义，张立勋，张今瑜等.1R2T绳索牵引骨盆康复机器人动力学.江苏大学学报：自然科学版.2010,31(2):131-135页
[67]王克义，张立勋，孟浩.1R2T绳索牵引并联康复机器人绳索弹性研究.南京理工大学学报:自然科学版.2010,34(5):602-607页
[68]张立勋，王克义，张今瑜等.基于绳索牵引的骨盆运动并联康复机器人的可控性研究.哈尔滨工程大学学报.2007,28(7):790-794页
[69]刘攀，张立勋，王克义等.绳牵引康复机器人的动力学分析与控制.中国机械工程.2009,20(11):1335-1339页
[70]张立勋，刘攀，王克义等.三自由度绳牵引平面并联机器人力/位控制研究.机械设计与研究.2008,24(3):22-25页
[71]张立勋，刘攀，王克义等.绳索牵引并联康复机器人的建模与控制.哈尔滨工程大学学报.2009,30(1):81-85页
[72]张琦，李琪，纪树荣.骨盆训练对脑卒中偏瘫患者步行能力的影响.中国康复理论与实践.2001,7(4):161-162页
[73]陈东辉，佟金，李重焕等.人和动物的步态与步行机器人.吉林大学学报.2003,33(4):121-125页
[74]藏红.人体运动学.人民卫生出版社,2008:20-26页
[75] D A Winter, C D Mackinnon, G K Ruder, et al. An integrated EMG biomechanical modelof upper body balance and posture during human gait. Progress in Brain Research.Amsterdam:Elsevier.1993,(97)113-156P
[76] C D Mackinnon, D A Winter. Control of whole body balance in the frontal plane duringhuman walking. Journal of Biomechanics.1993,26(6):33-44P
[77] ZHAO Lingyan, ZHANG Lixun, WANG Lan, et al. Three-dimensional motion of thepelvis during human walking. In:IEEE ICMA2005. Niagara Falls.2005:335-339P
[78] Y Stauffer. Pelvic Motion Implementation on the WalkTrainer.2007IEEE InternationalConference on Robotics and Biomimetics. Sanya,2007:133-138P
[79]古恩鹏，刘爱峰，金鸿宾等.步态分析在临床骨科与康复中的应用.中国中西医结合外科杂志.2011,17(3):335-337页
[80]王桂茂，严隽陶，刘玉超等.脑卒中偏瘫步态的时空参数与骨盆运动学分析.临床研究.2010,25(12):1148-1151页
[81] F M L Amirouche, S K Ider, J Trimble. General Analytical method for the analysis andsimulation of human locomotion. American Society of Mechanical Engineers,Bioengineering Division (Publication) BED.1988:57-60P
[82]方海峰.人体下肢有限元动力学分析模型的建立和验证.湖南大学硕士论文.2004:22页
[83] http://www.entdr.com.cn/Get/jichushuji02/163133813.htm
[84] http://www.hudong.com/wiki/%e9%ab%8b%e5%85%b3%e8%8a%82
[85]张更林，金宝士，吴雅杰.人体下肢机械式膝关节运动分析.中国临床康复.2004,8(2):210-212页
[86] W Zijlstra, A L Hof. Displacement of the pelvis during human walking: experimentaldata and model predictions. Gait and Posture.1997,(8):249-267P
[87]朱镛连.神经康复学.北京人民军医出版社,2001:154-166页
[88] T G Homby, D H Zemon, D Campbell. Robotic-assisted, body-weight-supportedtreadmill training in individuals following motor incomplete spinal cord injury. PhysicalTherapy,2005,85(1):52-66P
[89] Wang Qiyuan, Qian Jinwu, Zhang Yanan, et al. Gait Trajectory Planning and Simulationfor the Powered Gait Orthosis.2007IEEE International Conference on Robotics andBiomimetics. China: Yalong Bay, Sanya,2008:1693-1697P
[90]毕秋吉.步态训练机器人骨盆控制机构研究.哈尔滨工程大学硕士论文.2009:18-21页
[91]夏振涛.助行训练机器人系统设计及步态控制实验研究.哈尔滨工程大学硕士论文.2009:32-35页
[92]王伟.绳驱动助行机构控制.哈尔滨工程大学.2009:38-40页
[93]赵豫玉.穿戴式下肢康复机器人的研究.哈尔滨工程大学硕士论文.2009:22-24页
[94] Y Y Chew, S N Mark, N C John. Automated person recognition by walking and runningvia model-based approaches. Pattern Recognition.2004(37):1057-1072P
[95] G S Anthony, B Peter, R David. Three-dimensional angular kinematics of the lumbarspine and pelvis during running. Human Movement Science.2002(21):273-293P
[96] P H Jason, J P Anthony, J R Paul. Explicit finite element modeling of total kneereplacement mechanics. Journal of Biomechanics.2005(38):323-331P.
[97]宋伟刚.机器人学—运动学、动力学与控制.北京:科学出版社,2007:81-87页
[98]万晓正.平面绳牵引并联机构的控制研究.哈尔滨工程大学硕士论文.2009:33-36页
[99]董玉红.欠驱动力合作机器人的驱动及控制技术研究.哈尔滨工程大学博士论文.2006:42-55页
[100] J.H.金斯伯格著.白化同，李俊宝译.机械与结构振动—理论与应用.北京:中国宇航出版社,2005:12-13,22-23页
[101] Furuta, Hara, Mori. A class of systems with the same observer. IEEE Trans. Auto.Contr,1976(8):572-576P
[102] A RGralimidi, B R Barmish. The constrained Lyapunov problem and its aplication torobust output feedback stabilization. IEEE Trans.Auto.Contr.1986,31(5):410-419P
[103]梅生伟，申铁龙，刘康志.现代鲁棒控制理论与应用.北京:清华大学出版社,2003:19-22页
[104]储健，俞立，苏宏业.鲁棒控制理论及应用.杭州:浙江大学出版社,2000:55-58页
[105]俞立.鲁棒控制—线性矩阵不等式处理方法.北京:清华大学出版社,2002:18-19页
[106]俞立.鲁棒控制—线性矩阵不等式处理方法.北京:清华大学出版社,2002:27-28页
[107] L R Petersen. Quadratic stabilizability of uncertain linear systems containing bothconstant and time-varying uncertain parameters. Journal of Optimization Theory andApplications.1988,57(3):439-461P
[108] F Jabbari, W E Schmitendorf. Robust linear controllers using observers. IEE Trans AutoContr.1991,36(12):1509-1514P
[109]朱晓东，孙优贤.不确定动态时滞系统的基于观测器的鲁棒镇定设计.控制理论与应用.1996,13(2):254-257页
[110]苏宏业，王景成，褚健.一类不确定动态时滞系统的无记忆鲁棒镇定控制.自动化学报.1997,24(4):497-501页