智能仿生手臂肌电信号—运动模型化与模式识别理论方法研究
|
摘要
智能仿生手臂指能够根据人的大脑发出的意念自由地控制假手完成各种动作，在抓取物体时能够感知物体的外部特征，自动调整假手系统的参数，使假手完成灵活、可靠的动作的高性能新一代假肢。目前用于智能仿生手臂的控制信号有很多，其中表面肌电信号易于获取且控制假肢具有直接、自然的特点，已成为应用最为广泛的控制信号源。表面肌电信号的处理质量直接关系着智能仿生手臂控制的安全性与可靠性，本文主要针对表面肌电信号进行分析处理。由于其具有信号微弱、易受噪声干扰、个体差异性大等特点，为提高多运动模式识别率，本文主要进行了以下几个方面的研究：
寻找肌肉与不同运动模式间的对应关系，确定最佳肌电电极摆放位置，检测合理活动段。人体肌肉与运动的关系较为复杂，每一块肌肉所对应的功能不尽相同，且相互影响。表面肌电信号采集效果受电极位置影响较大，准确的采集肌电信号使其能清晰的反映不同运动模式下肌肉的运动情况是肌电信号处理的前提。本文借助人体解剖学、生物医学等知识选择手臂指浅屈肌和桡侧腕屈肌作为电极摆放位置，降低了信号采集的随机性，保证了信号的准确性。文中利用MQ-8型表面肌电采集系统，对8位健康测试者采集6种运动模式的表面肌电信号，并用移动平均值法处理表面肌电信号序列的瞬时能量，通过结合阈值进行比较，最终确定信号的有效活动段。经过上述简单必要的预处理过程，找到信号中表示动作起始和终止的位置，即检测到有效活动段，有助于后续表面肌电信号分析。
提出非抽样双密度小波变换和双参数软阈值去噪的方法。表面肌电信号在采集过程中噪声无法避免，虽然通过硬件采集系统可去除部分噪声，使有用信号不至于被噪声淹没，但在信号传输过程中仍存在大量仪器噪声。针对此情况，本文采用非抽样双密度小波变换对信号进行分析，此方法具有严格平移不变性，避免了信号重构过程中人为噪声的引入，且与离散小波变换相比，具有两个小波函数，能够在同一尺度内多获得一组信号细节信息，有助于表面肌电信号精细去噪。采用双参数软阈值法对分解的信号高频部分去噪，与传统软阈值方法相比，此方法便于调节，在阈值附近具有较为平滑的过渡带，符合自然信号连续特性，能克服软阈值函数存在的缺陷。特别在小于阈值部分，没有将函数值至零，而是根据表面肌电信号特性，最大程度的保留信号细节信息，符合其高精准度去噪的要求。
提出基于傅里叶级数和FFT盲辨识的表面肌电信号建模方法。肌电信号建模是研究肌电信号特性的一种重要途径，目前常用的肌电信号数学模型有线性系统模型、非平稳模型、双极型模型以及集中参数模型4种。以上肌电信号建模的方法从不同方面描述肌电信号与肌肉力的关系，提供了判断肌肉活动状况的理论依据，被广泛应用于临床医学、康复医学与工程及运动医学和运动生物力学领域。上述建模方法着重从生物医学方面描述肌电信号，有助于判断运动疲劳、肌肉异常等状况，无法表现肌电信号与不同运动模式之间的关系，很少在仿生假肢领域使用。针对此情况，文中尝试采用傅里叶级数法和FFT盲辨识法分别建立肌电信号模型，寻找信号与运动间的关系，并采用模型系数作为信号特征，通过实验可见，本文方法建立的模型系数对多运动模式的表面肌电信号具有较好的表征能力，将其作为信号特征有助于后续多运动模式识别。
提出半监督boosting多模式识别方法。由于标注样本较难获取，且相似运动模式的表面肌电信号较难区分，为获得较高模式识别率，对分类器性能要求较高。针对此情况，文中采用半监督学习法，其与监督学习和无监督学习相比，可通过少量标注样本和大量无标注样本训练基分类器，使其具有较高的识别能力。且本文方法与现有半监督方法主要针对两分类问题不同，其可直接处理多类问题，由于无需将多类问题转化为多个两类问题，避免了分类结果处于多个尺度无法相加以及将某一样本划分为多类的情况。Boosting算法是半监督学习领域较常用的一种方法，其可被认为是一种自我训练过程，通过某一框架结构联合多个现有分类器，可实现提高分类性能的目的。与现有半监督boosting仅依据分类置信度调整无标注样本的伪类别不同，本文方法将分类置信度与样本间相似度相结合共同作为无标注样本伪类别的调整依据，其最大程度的有效利用了多种假设与聚类假设。实验证明，本文算法对不同基分类器、不同标注样本数均具有较强的鲁棒性。
以表面肌电信号模式识别结果作为控制信号，采用自抗扰方法控制Adams虚拟手臂完成指定动作，验证肌电信号手臂控制的准确性和稳定性。利用Adams仿真软件设计开发具有16自由度假手，并依据此建立其动力学模型，采用自抗扰控制器调整参考输入与Adams实际输出间的误差，形成稳定的闭环控制。自抗扰控制器由PID控制器演变而来，其采用了PID误差反馈控制的核心理念。该方法不依赖于被控对象精确的数学模型，算法简单，在未知强非线性和不确定强扰动作用下能保证控制精度，适用于具有较高自由度的仿生手臂控制。
最后，总结了全文所做的工作，提出了今后进一步需要研究的问题。
According to the ideas issued by brain, intelligent bionic arm can control the prosthetichand to complete a variety of actions freedomly. The new generation prosthetic hand cansense the external characteristics of the objects, adjust the parameters of the hand systemautomatically and complete proposed action flexibly and reliably. At present, there are lots ofintelligent bionic arm control signals. Surface eletromyographic signal (sEMG) obtain easilyand control directly and naturally which has proven to be an effective control input forpowered upper limb prostheses. The sEMG signal processing quality influences the securityand reliability of control, directly. This paper mainly aims at using sEMG processing toextract more information from the input signals to increase the accuracy of the controllers. Itsmain contents are presented as follows:
Looking for the corresponding relationship between muscle and different movementmodes，identify the best position for putting myoelectricity electrode, testing for reasonableactivities section. The relationship between movements and human muscles is complicated,the corresponding function of each muscle is different, and they influence each other. Theposition of electrode might have impact during collecting EMG. It is the premise of EMGsignals processing that accurate collecting EMG signals to reflect muscle movements underdifferent movement patterns correctly. In this paper, the musculus flexor digitorum sublimisand the musculus flexor carpi radialis are chosen as the position for putting electrode throughhuman anatomy and biological medicine, which can reduce the randomness of the signalcollection and guarantee the accuracy of the signal. In this paper, the sEMG of six movementmotions for eight healthy testers are collected using MQ-8sEMG collection system. Thenmoving average method is used to process instantaneous energy of sEMG sequence. Finallythe signal effectively activities are determined through the comparison with threshold. Afterthe above pretreatment process, the effective activities section is detected, which can help thesEMG analysis in the subsequence.
Undecimated double density wavelet transform (UDDWT) and double-parameter softthreshold are proposed to meet the requirement of sEMG nondestructinve de-noisnig. Although the hardware system can remove a part of noise, which made the desired signal ofsEMG not be drowned. It is sill existente during the acquisition and signal transmissionprocess, which have a lot of instrumental noise. In view of this situation, the undecimateddouble density wavelet analysis of signal is used in this paper. It has strict shift invariance,which can avoid the artifacts in signal reconstruction. UDDWT has two wavelet functionswhich can get more detail in every scale compared with discrete wavelet transform (DWT).Double-parameter soft threshold method, which has a relatively smooth transition zone, isused for denoising the high frequency of the signal in the signal decomposition. It cansuppress the noise efficiently meanwhile keep useful detail information to the greatest extent.Experiment results show that the proposed methods, which get high signal-to-ratio whileretaining the signal characteristics, are very suitable for multi-channel sEMG of similar handmovements de-noising.
Fourier series and FFT-based blind identification methods are proposed to establishsEMG model. sEMG signal modeling has become an important approach to study of EMGsignal characteristics. Now there are four methds are commonly used in sEMG signalmathematical model, which are linear system model, nonstationary model, bipolar model andthe lumped parameter model respectively. The methods mentioned above, usually describethe relationship between EMG and muscle force from the different aspects to provide thetheoretical basis for judging of muscle activity status, have been widely applied in clinicalmedicine, rehabilitation medicine, sports medicine, neurophysiology and ergonomics. Thosemodeling methods are described emphatically from the biomedical aspects of EMG signals,which can help to distinguish the motion fatigue or muscle abnormalities. They cannot reflectto the relationship between EMG signals with different movement patterns, therefore they arerarely used in intelligent bionics. In these circumstances, Fourier series method and FFT blindidentification method are proposed to build EMG model in this paper to explore relationshipbetween signals and different motions. The model coefficients are used as the signalcharacteristics, the method proposed in this paper has good characterization capabilitiesthrough experiments. By using these as the signal features contribute to subsequentmovement pattern recognition.
Semi-supervised boosting algorithm is proposed to classify different movement motions.Traditionally, machine learning is categorized as two paradigms i.e. supervised versusunsurpervised learning. Supervised learning finds out a rule for the predictive relationshipbetween input and output from a set of finite examples in the format of input-output pairs,while unsupervised learning seeks a structure of interests underlying a data set. In general, supervised learning requires many training examples to establish a learner of the satisfactorygeneralization capability. The acquisition of training examples is nontrivial for supervisedlearning, which needs to annotate input data with appropriate labels. In sEMG, the labeledinput data is often difficult, expensive, and time-consuming, especially when it has to be donemanually by experts. On the other hand, there is often a massive amount of unlabeled dataavailable. In order to exploit unlabeled data, semi-supervised learning has become a novelparadigm by using a large number of unlabeled points together with a small number oflabeled examples to build a better learner. Most semi-supervised learning algorithms havebeen designed for binary classification, and are extended to multi-class classification byapproaches such as one-against-the-rest. The main shortcoming of these approaches is thatthey are unable to exploit the fact that each example is only assigned to one class. Additionalproblems with extending semi-supervised binary classifiers to multi-class problems includeimbalanced classification and different output scales of different binary classifiers. A newsemi-supervised boosting algorithm is proposed in this paper that directly solves thesemi-supervied multi-class learning problem. Compared to the existing semi-supervisedboosting methods, the proposed algorithm is advantageous in that it exploits bothclassification confidence and similarities among examples when deciding the pseudo-labelsfor unlabeled examples. The experiments show that the proposed algorithm performs betterthan the state-of-the-art boosting algorithms for semi-supervised learning.
Active disturbance rejection control (ADRC) develops the kenel of the PID control, andcombines the observer of modern control ideas. It is a new control method, which algorithmis simple and adaptability, and can automatically compensate internal and externaldisturbances. It can be widely used in industrial applications. ADRC is introduced into bionicaritficail arm in this paper. Adams is used to design artificial arm with16DOF. Themathematical model is obtained from the established artificial arm by using kinematics anddynamics analysis. ADRC adjusts the error between reference input and the Adams actualoutput, to form a stable closed loop control, according to the mathematical model motionedabove. The control method does not depent on the system model, and has the strongadaptation, robusness and operability.
Finally, the main content of this dissertation is summarized, and the further researchesare discussed.
寻找肌肉与不同运动模式间的对应关系，确定最佳肌电电极摆放位置，检测合理活动段。人体肌肉与运动的关系较为复杂，每一块肌肉所对应的功能不尽相同，且相互影响。表面肌电信号采集效果受电极位置影响较大，准确的采集肌电信号使其能清晰的反映不同运动模式下肌肉的运动情况是肌电信号处理的前提。本文借助人体解剖学、生物医学等知识选择手臂指浅屈肌和桡侧腕屈肌作为电极摆放位置，降低了信号采集的随机性，保证了信号的准确性。文中利用MQ-8型表面肌电采集系统，对8位健康测试者采集6种运动模式的表面肌电信号，并用移动平均值法处理表面肌电信号序列的瞬时能量，通过结合阈值进行比较，最终确定信号的有效活动段。经过上述简单必要的预处理过程，找到信号中表示动作起始和终止的位置，即检测到有效活动段，有助于后续表面肌电信号分析。
提出非抽样双密度小波变换和双参数软阈值去噪的方法。表面肌电信号在采集过程中噪声无法避免，虽然通过硬件采集系统可去除部分噪声，使有用信号不至于被噪声淹没，但在信号传输过程中仍存在大量仪器噪声。针对此情况，本文采用非抽样双密度小波变换对信号进行分析，此方法具有严格平移不变性，避免了信号重构过程中人为噪声的引入，且与离散小波变换相比，具有两个小波函数，能够在同一尺度内多获得一组信号细节信息，有助于表面肌电信号精细去噪。采用双参数软阈值法对分解的信号高频部分去噪，与传统软阈值方法相比，此方法便于调节，在阈值附近具有较为平滑的过渡带，符合自然信号连续特性，能克服软阈值函数存在的缺陷。特别在小于阈值部分，没有将函数值至零，而是根据表面肌电信号特性，最大程度的保留信号细节信息，符合其高精准度去噪的要求。
提出基于傅里叶级数和FFT盲辨识的表面肌电信号建模方法。肌电信号建模是研究肌电信号特性的一种重要途径，目前常用的肌电信号数学模型有线性系统模型、非平稳模型、双极型模型以及集中参数模型4种。以上肌电信号建模的方法从不同方面描述肌电信号与肌肉力的关系，提供了判断肌肉活动状况的理论依据，被广泛应用于临床医学、康复医学与工程及运动医学和运动生物力学领域。上述建模方法着重从生物医学方面描述肌电信号，有助于判断运动疲劳、肌肉异常等状况，无法表现肌电信号与不同运动模式之间的关系，很少在仿生假肢领域使用。针对此情况，文中尝试采用傅里叶级数法和FFT盲辨识法分别建立肌电信号模型，寻找信号与运动间的关系，并采用模型系数作为信号特征，通过实验可见，本文方法建立的模型系数对多运动模式的表面肌电信号具有较好的表征能力，将其作为信号特征有助于后续多运动模式识别。
提出半监督boosting多模式识别方法。由于标注样本较难获取，且相似运动模式的表面肌电信号较难区分，为获得较高模式识别率，对分类器性能要求较高。针对此情况，文中采用半监督学习法，其与监督学习和无监督学习相比，可通过少量标注样本和大量无标注样本训练基分类器，使其具有较高的识别能力。且本文方法与现有半监督方法主要针对两分类问题不同，其可直接处理多类问题，由于无需将多类问题转化为多个两类问题，避免了分类结果处于多个尺度无法相加以及将某一样本划分为多类的情况。Boosting算法是半监督学习领域较常用的一种方法，其可被认为是一种自我训练过程，通过某一框架结构联合多个现有分类器，可实现提高分类性能的目的。与现有半监督boosting仅依据分类置信度调整无标注样本的伪类别不同，本文方法将分类置信度与样本间相似度相结合共同作为无标注样本伪类别的调整依据，其最大程度的有效利用了多种假设与聚类假设。实验证明，本文算法对不同基分类器、不同标注样本数均具有较强的鲁棒性。
以表面肌电信号模式识别结果作为控制信号，采用自抗扰方法控制Adams虚拟手臂完成指定动作，验证肌电信号手臂控制的准确性和稳定性。利用Adams仿真软件设计开发具有16自由度假手，并依据此建立其动力学模型，采用自抗扰控制器调整参考输入与Adams实际输出间的误差，形成稳定的闭环控制。自抗扰控制器由PID控制器演变而来，其采用了PID误差反馈控制的核心理念。该方法不依赖于被控对象精确的数学模型，算法简单，在未知强非线性和不确定强扰动作用下能保证控制精度，适用于具有较高自由度的仿生手臂控制。
最后，总结了全文所做的工作，提出了今后进一步需要研究的问题。
According to the ideas issued by brain, intelligent bionic arm can control the prosthetichand to complete a variety of actions freedomly. The new generation prosthetic hand cansense the external characteristics of the objects, adjust the parameters of the hand systemautomatically and complete proposed action flexibly and reliably. At present, there are lots ofintelligent bionic arm control signals. Surface eletromyographic signal (sEMG) obtain easilyand control directly and naturally which has proven to be an effective control input forpowered upper limb prostheses. The sEMG signal processing quality influences the securityand reliability of control, directly. This paper mainly aims at using sEMG processing toextract more information from the input signals to increase the accuracy of the controllers. Itsmain contents are presented as follows:
Looking for the corresponding relationship between muscle and different movementmodes，identify the best position for putting myoelectricity electrode, testing for reasonableactivities section. The relationship between movements and human muscles is complicated,the corresponding function of each muscle is different, and they influence each other. Theposition of electrode might have impact during collecting EMG. It is the premise of EMGsignals processing that accurate collecting EMG signals to reflect muscle movements underdifferent movement patterns correctly. In this paper, the musculus flexor digitorum sublimisand the musculus flexor carpi radialis are chosen as the position for putting electrode throughhuman anatomy and biological medicine, which can reduce the randomness of the signalcollection and guarantee the accuracy of the signal. In this paper, the sEMG of six movementmotions for eight healthy testers are collected using MQ-8sEMG collection system. Thenmoving average method is used to process instantaneous energy of sEMG sequence. Finallythe signal effectively activities are determined through the comparison with threshold. Afterthe above pretreatment process, the effective activities section is detected, which can help thesEMG analysis in the subsequence.
Undecimated double density wavelet transform (UDDWT) and double-parameter softthreshold are proposed to meet the requirement of sEMG nondestructinve de-noisnig. Although the hardware system can remove a part of noise, which made the desired signal ofsEMG not be drowned. It is sill existente during the acquisition and signal transmissionprocess, which have a lot of instrumental noise. In view of this situation, the undecimateddouble density wavelet analysis of signal is used in this paper. It has strict shift invariance,which can avoid the artifacts in signal reconstruction. UDDWT has two wavelet functionswhich can get more detail in every scale compared with discrete wavelet transform (DWT).Double-parameter soft threshold method, which has a relatively smooth transition zone, isused for denoising the high frequency of the signal in the signal decomposition. It cansuppress the noise efficiently meanwhile keep useful detail information to the greatest extent.Experiment results show that the proposed methods, which get high signal-to-ratio whileretaining the signal characteristics, are very suitable for multi-channel sEMG of similar handmovements de-noising.
Fourier series and FFT-based blind identification methods are proposed to establishsEMG model. sEMG signal modeling has become an important approach to study of EMGsignal characteristics. Now there are four methds are commonly used in sEMG signalmathematical model, which are linear system model, nonstationary model, bipolar model andthe lumped parameter model respectively. The methods mentioned above, usually describethe relationship between EMG and muscle force from the different aspects to provide thetheoretical basis for judging of muscle activity status, have been widely applied in clinicalmedicine, rehabilitation medicine, sports medicine, neurophysiology and ergonomics. Thosemodeling methods are described emphatically from the biomedical aspects of EMG signals,which can help to distinguish the motion fatigue or muscle abnormalities. They cannot reflectto the relationship between EMG signals with different movement patterns, therefore they arerarely used in intelligent bionics. In these circumstances, Fourier series method and FFT blindidentification method are proposed to build EMG model in this paper to explore relationshipbetween signals and different motions. The model coefficients are used as the signalcharacteristics, the method proposed in this paper has good characterization capabilitiesthrough experiments. By using these as the signal features contribute to subsequentmovement pattern recognition.
Semi-supervised boosting algorithm is proposed to classify different movement motions.Traditionally, machine learning is categorized as two paradigms i.e. supervised versusunsurpervised learning. Supervised learning finds out a rule for the predictive relationshipbetween input and output from a set of finite examples in the format of input-output pairs,while unsupervised learning seeks a structure of interests underlying a data set. In general, supervised learning requires many training examples to establish a learner of the satisfactorygeneralization capability. The acquisition of training examples is nontrivial for supervisedlearning, which needs to annotate input data with appropriate labels. In sEMG, the labeledinput data is often difficult, expensive, and time-consuming, especially when it has to be donemanually by experts. On the other hand, there is often a massive amount of unlabeled dataavailable. In order to exploit unlabeled data, semi-supervised learning has become a novelparadigm by using a large number of unlabeled points together with a small number oflabeled examples to build a better learner. Most semi-supervised learning algorithms havebeen designed for binary classification, and are extended to multi-class classification byapproaches such as one-against-the-rest. The main shortcoming of these approaches is thatthey are unable to exploit the fact that each example is only assigned to one class. Additionalproblems with extending semi-supervised binary classifiers to multi-class problems includeimbalanced classification and different output scales of different binary classifiers. A newsemi-supervised boosting algorithm is proposed in this paper that directly solves thesemi-supervied multi-class learning problem. Compared to the existing semi-supervisedboosting methods, the proposed algorithm is advantageous in that it exploits bothclassification confidence and similarities among examples when deciding the pseudo-labelsfor unlabeled examples. The experiments show that the proposed algorithm performs betterthan the state-of-the-art boosting algorithms for semi-supervised learning.
Active disturbance rejection control (ADRC) develops the kenel of the PID control, andcombines the observer of modern control ideas. It is a new control method, which algorithmis simple and adaptability, and can automatically compensate internal and externaldisturbances. It can be widely used in industrial applications. ADRC is introduced into bionicaritficail arm in this paper. Adams is used to design artificial arm with16DOF. Themathematical model is obtained from the established artificial arm by using kinematics anddynamics analysis. ADRC adjusts the error between reference input and the Adams actualoutput, to form a stable closed loop control, according to the mathematical model motionedabove. The control method does not depent on the system model, and has the strongadaptation, robusness and operability.
Finally, the main content of this dissertation is summarized, and the further researchesare discussed.
引文
[1]BRODY G, BALASUBRAMANIAN R, SCOTT R.N. A model for myoelectrical signalgeneration[J]. MED. BIOL. ENGINEERING.1974,12:29-41.
[2]EDUARDO MANGIERI et al. A Novel Analogue Circuit for Controlling ProstheticHands[C]. IEEE Biomedical Circuits and Systems Conference, USA,2008:81-84.
[3]MERLETTI R., RAINOLDI A., FARIAN D. Surface electromyography for noninvasivecharacterization of muscle[J]. Exercise and Sport Sciences Reviews.2001,29:20-25.
[4] BICCHI A. Hand for Dexterous Manipulation and Robust Grasping: A Difficult RoadToward Simplicity[J]. IEEE Trans. Robot. Autom.2000,16(6):652-662.
[5] BIAGIOTTI L., LOTTI F., MELCHIORRI C., VASSURA G.. How Far Is the HumanHand? A Review on Anthroponorphic Robotic End-effectors. Italy: University ofBologna.[2009-04-05].http://www-lar.deis.upioe.it/woda/data/deis-lar-publications/3cbd.Document.pdf
[6] DASHY R., VEI C., LABIANC M.. The Design and Development of a GlovelessEndoskeleton of Prosthetic Hand[J]. Journal of Rehabilitation Research and Development.1998,35(4):388-395.
[7] DECHEV N., LEGHORN W. L., NEUMANN S.. Multiple Finger Passive AdaptiveGrasp Prosthetic Hand[J]. Mechanism and Machine Theory.2001,36:1157-1173.
[8] KYBERD P. J., LIGHT C. M., CHAPPEL P. H., et al. The Design of AnthropomorphicProsthetic Hands: A Study of the Southampton Hand[J]. Robotica.2001,19(6):593-600.
[9] LIGHT C. M., CHAPPELL P. H.. Development of a lightweight and adaptablemultiple-axis hand prosthesis[J]. Medical Engineering and Physics.2000,22(10):679-684.
[10] KYBERD P. J.. The intelligent hand[J]. IEEE Review.2000,9:31-35.
[11] ZHIXIU HAO, YONGSHENG SHEN. Force self-adaptive control for prehensors[J]. JTsinghua Univ (Sci&Tech),1999,39:92-99.
[12] HIRAIWA A., SHIMOHARA K., TOKUNAGA Y.. EMG Pattern Analysis andClassification by Neural Network[C]. IEEE International Conference on Systems, Man andCybernetics, Cambridge, MA, USA,1989:1113-1115.
[13] LOTTI F., TIEZZI P., MELCHIORRI C.. Development of UB Hand3[C]. Proceedingsof the2005IEEE International Conference on Robotics and Automation, Barcelona, Spain,2005:4499-4503.
[14] ZOLLO L., ROCCELLA S., GUGLIELMELLI E., et al. Biomechatronic Design andControl of an Anthropomorphic Artificial Hand for Prosthetic and Robotic Applications[J].IEEE/ASME Transactions on Mechatronics.2007,12(4):418-429.
[15] PYLATIUK C., MOUNIER S., KARGOV A., et al. Progress in the Development of aMultifunctional Hand Prosthesis[C]. The26thIEEE Annual International Conference onEngineering in Medicine and Biology Society, San Francisco, USA.2004:4260-4263.
[16] The i-limb system[Z].[2010-5-27]. www.touchbionics.com
[17] Otto Bock Michelangelo hand[Z].[2009-03-27]. http://www.swisswuff.ch/tech/?p=145.
[18] HE. HUANG, PING. ZHOU, et al. An Analysis of EMG Electrode Configuration forTargeted Muscle Reinnervation Based Neural Machine Interface[J]. IEEE Trans. NeuralSystems and Rehabilitation Engineering.2008,16(1):37-46.
[19] KUIKEN T. A., LEE G., LOCK B. A., et al. Targeted Muscle Reinnervation forReal-time Myoelectric Control of Multifuction Artificial Arms[J]. Journal of the AmericanMedical Association.2009,301(6):619-628.
[20] DALLEY S. A., WISTE T. E., WITHROW T. J., et al. Design of a MultifuctionalAnthropomorphic Prosthetic Hand With Extrinsic Actuation[J]. IEEE Transactions onMechatronics.2009,14(6):699-706.
[21]杨大鹏.特定心理作业多通道脑电信号识别系统研究[D].哈尔滨工业大学硕士学位论文.2006.
[22] ENGEL A. K., MOLL C. K. E., FRIED I., et al. Invasive Recordings from the HumanBrain: Clinical Insights and Beyond[J]. NATURE.2005,6:35-47.
[23] MEEL V., SAGI P., SPALDING C. M., et al. Cortical Control of A Prosthetic Arm forSelf-Feeding[J]. NATURE.2008,452(19):1098-1101.
[24] ZARDOSHTI K. M., WHEELER B. C., BADIE K., HASHEMI R. M. EMG FeatureSelection for Movement Control of a Cybernetic Arm[J]. Cybernetics and Systems.1995,26(2):189-210.
[25] BOOSTANI R., MORADI M. H. Evaluation of the Forearm EMG Signal Feature for theControl of a Prosthetic Hand[J]. Physiological Measurement.2003,24(2):309-319.
[26] KANG W. J., CHENG C. K., LAI J. S., SHIU J. R., KUO T. S. A Comparative Analysisof Various EMG Pattern Recognition Methods[J]. Medical Engineering and Physics.1996,18(5):390-395.
[27] KIM W. S., YOON Y. Z., BAE J. H., SOH K. S. Nonlinear Characteristics of Heart RateTime Series: Influence of Three Recumbent Positions in Patients With Mild or SevereCoronary Artery Disease[J]. Physiological Measurement.2005,26(4):517-529.
[28] INBAR G. F., ALLINE J., PAISS O., KRANZ H. Monitoring Surface EMG SpectralChanges By the Zero Crossing Rate[J]. Medical and Biological Engineering andComputing.1986,24(1):10-18.
[29] MERLETTI R., LOCONTE L. R., ORIZIO C. Indices of Muscle Fatigue[J]. Journal ofElectromyography and Kinesiology.1991,1(1):20-33.
[30] FARIAN D, MERLETTO R. Comparison of Algorithms for Estimation of EMGVariables During Voluntary Isometric Contractions[J]. Journal of Electromyography andKinesiology.2000,10(5):337-349.
[31] UCHIDA N., HIRAIWA A., SONEHARA N., SHIMOHARAL K.. EMG patternrecognition by neural networks for multi finger control[C]. Proc. Ann. Eng. Med. Biol.Soc.,1992:1016–1018.
[32] RONAGER J., CHRISTENSEN H. Power Spectrum Analysis of the EMG Pattern inNormal and Diseased Muscles[J]. Journal of the Neurologicla Sciences.1989,94(1-3):283-294.
[33] BILODIAU M., CINCERA M., GERVAIS S., et al. Changes in the ElectromyographicSpectrum Power Distribution Caused by a Progressive Increase in the Force Level[J].European Journal of Applied Physiology and Occupational Physiology.1995,71(2-3):113-123.
[34] ECKARDT L., HARZER W. Facial Structure and Functional Findings in Patients withProgressive Muscular Dystrophy(Duchenne)[J]. American Journal of Orthodontics andDentofacial orthopedics: Official Publication of the American Association of Orthodontists,Its Constituent Societies, and the American Board of Orthodontics.1996,110(2):185-190.
[35] FARINA D. Interpretation of the Surface Eletromyogram in Dynamic Contractions[J].Exercise and Sport Sciences Reviews.2006,34(3):121-127.
[36]王飞,罗志增.基于功率谱的表面EMG信号运动模式识别[J].杭州电子科技大学学报.2005,25(2):37—40.
[37] DAVIES M. R., REISMAN S. S. Time Frequency Analysis of the ElectromyogramDuring Fatigue[C]. Bioengineering Proceedings of the Northeast Conference, Springfield,MA, USA.1994
[38] JANG G. C., CHENG C. K., LAI J. S., KUO T. S. Using Time-Frequency AnalysisTechnique in the Classification of Surface EMG Signals[C]. IEEE Engineering inMedicine and Biology Proceedings, Baltimore, MD, USA.1994.
[39] BONATO P., GAGLIATI G., KNAFLITZ M. Analysis of Myoelectric Signals RecordedDuring Dynamic Contractions[J]. IEEE Engineering in Medicine and Biology Magazine.1996,15(6):102-111.
[40] BECK T. W., HOUSH T. J., JOHNSON G. O., WEIR J. P., et al. Comparison of Fourierand Wavelet Transform Procedures for Examining the Mechanomyographic andElectromyographic Frequency Domain Response During Fatiguing Isokinetic MuscleActions of the Biceps Brachii[J]. Journal of Electromyography and Kinesiology.2005,15(2):190-199.
[41] MERLO A., FARINA D., MERLETTI R. A Fast and Reliable Technique for MuscleActivity Detection From Surface EMG Signals[J]. IEEE Transactions on BiomedicalEngineering.2003,50(3):316-323.
[42] KARLSSON S., YU J., AKAY M. Time-Frequency Analysis of Myoelectric SignalsDuring Dynamic Contractions: A Comparative Study[J]. IEEE Transactions on BiomeicalEngineering.2000,47(2):228-238.
[43] HOBO XIE, ZHIZHONG Wang. Mean Frequency Derived Via Hilbert-HuangTransform with Application to Fatigue EMG Signal Analysis[J]. Computer Methods andPrograms in Biomedicine.2006,82(2):114-120.
[44] CHAN F. H. Y., YANG Y. S., LAM F. K., ZHANG Y. T., PARKER P. A.. Fuzzy EMGClassification for Prosthesis Control[J]. IEEE Trans. Rehabil. Eng.,2000,8(3):305-311.
[45] PINGAO MEI, ZHIZENG LUO. sEMG Disposal Based on Wavelet Packet Analysis andElman Neural Network[J]. Mechanical&Electrical Engineering Magazine,2008,25:7-10.
[46] YIHUNG LIU, HANPANG HUANG, CHANGHSIN WENG. Recognition ofElectromyographic Signals Using Cascaded Kernel Learning Machine[J]. IEEE/ASMETransactions on Mechatronics,12:253-264.
[47] ZHIHONG LIU, ZHIZENG LUO. Hand Motion Pattern Classifier Based on EMGUsing Wavelet Packet Transform and LVQ Neural Networks[C]. Proceedings of2008IEEE International Symposium on IT in Medicine and Education, China,2008:28-32.
[48] JINGDONG ZHAO, JIANG LI, MINGHE JIN, HONG LIU, HEGAO CAI. ProstheticHand Control Based on EMG Signal[J]. Machinary&Electron,2006,8:54-56.
[49] HAIHONG ZHANG, LIYU CAI, ZHIZHONG WANG. A Pi-Sigma neural networkbased electromyograph signal identification method[J]. Chinese Journal of RehabilitationMedicine.2000,15:34-37.
[50] XIANG CHEN, XU ZHANG, ZHANGYAN ZHAO, et al. Multiple Hand GestureRecognition based on Surface EMG Signal[C].1stInternatinal Conference onBioinformatics and Biomedical Engineering, China,2007:506—509.
[51] ZHIZENG LUO, FEI WANG. Electromyography Movement Pattern Recognition Basedon the Wavelet Eigenvalues and Pi-sigma Network[J]. Journal of Test And MeasurementTechnology,2006,20:344-348.
[52]李强.表面肌电信号的运动单位动作电位检测[D].中国科学技术大学博士论文.2008.
[53] CLANCY E. A., MORIN E. L., MERLETTI R. Sampling, Noise-Reduction andAmplitude estimation Issues in Surface Electromyography[J]. Journal ofElectromyography and Kinesiology.2002,12(1):1-16
[54] REAZ M. B. I., HUSSAIN M. S., MOHDYASIN F. Techniques of EMG Signal Analysis:Detection, Processing, Classification and Applications[J]. Biological Procedures Online.2006:11-35.
[55] PARKER. Signal Processing for Multistate Myoeletric Channel[J]. Proc. IEEE.1997,65(5):662-670.
[56] PARKER. Myoeletric Control of Prosthsis[J]. CRC Critical Review BiomedicalEngineering.1986,13(4):103-107.
[57] SHWEDYK. A Nonstationary Model for the Electromyogram[J]. IEEE Trans. BME-24.1997:417-429.
[58] CHEUNG J，HALBERTSMA J P，VELDHUIZEN A G. A Preliminary Study onElectromyographic Analysis of the Para Spinal Musculature in Idiopathic Scoliosis[J]. EurSpine J，2005，14(2):130-137.
[59] SHERIF M. H. Effects of Load on Myoeletric Signals: The ARIMA Representation[J].IEEE Trans. Biomed. Eng.1981,28:411-416.
[60] Brain-controlled'robo-arm'hope, BBC NEWS UK EDITION,Friday,18February,2005.
[61] KATOH R., HIROSHI Y., TAMIO A. Competitive Learning Method for RobustEMG-to-Motion Classifier[C]. IAS-9,2006:7-9.
[62] GRAUPE D. Multifunctional Prosthesis and Orthosis Control ViamicrocomputerIdentification of Temporal Pattern Differneces in Singlesite Myoeletric Signals[J]. Biomed.Eng.1982,4:17-22.
[63] JINGDONG ZHAO, ZONGWU XIE, HONG LIU, et al. Levenberg-Marquardt BasedNeural Network Control for a Five-fingered Prosthetic Hand[C]. IEEE Int. Conf.Roboticsand Automation, China,2005:4493–4498.
[64]汤小芙.临床肌电图学[M].北京：北京医科大学中国协和医科大学联合出版社.1995
[65]蔡立羽,王志中,刘瑜.基于盲辨识理论的双通道肌电信号建模与分类[J].上海交通大学学报.2000,34(11):1471-1474.
[66] DELUCA C. Physiology and Mathematics of Myoeletrics Signals[J]. IEEE Trans.Biomed. Eng.1979,26(6):313-325.
[67] MIN LEI, ZHIZHONG WANG, ZHENGJIN FENG. Detecting Nonlinearity of ActionSurface EMG Signal[J]. Physics Letters A.2001,290:297-303.
[68] BILODIAU M. Variations in the Shape of the EMG Power Specturm During anIncreasing Force Level[C]. Annual International Conference of the IEEE Engineering inMedicine&Biology Society-Proceedings.1993,15:1205-1206.
[69]何乐生.基于肌电信号的人机接口技术的研究[D].东南大学博士论文.2006.
[70] OSKOEI M., HU H. Myoelectric Control Systems-A Survey[J]. Biomedical SignalProceeding and Control.2007,2(4):275-294.
[71] SREE P S J, KUMAR P, SIDDAVATAM R, VERMA R. Slat-And-Pepper NoiseRemoval By Adaptive Median-Based Lifting Filter Using Second-Generation Wavelets[J].Signal, Image and Video Processing,2011:1-8(Article in Press)
[72] WEIXIA ZHAN, JIWEN TAN, YAN WEN. De-noising of Damage Signal in Wire RopeBased on Adaptive Lifting Wavelet[J]. Manufacturing Engineering and Automation,2010,139-141:2556-2560
[73] NONOMURA K, ONO M, ZHOU L B, SHIMIZU J, OJIMA H. Design of Digital FiltersFor Si Wafer Surface Profile Measurement-Noise Reduction By Lifting Scheme WaveletTransform-Noise[J]. Advances in Abrasive Technology XIII,2010,126-128:732-737
[74] INOUE H, MIYAZAKI A. Noise Reduction Method For ECG Signals Using The DyadicWavelet Transform[J]. IEICE Transactions on Fundamentals of ElectronicsCommunications and Computer Sciences,1998, E81-A(6):1001-1007
[75]祝海龙，郭天佑，屈梁生.基于二进小波变换和软阈值改进的信号消噪[J].自动化学报，2004,30(2):199-206
[76] SETHU V, AMBIKAIRAJAH E, Ge L. Selective Weighting Of Undecimated WaveletCoefficients For Noise Reduction In SAR Interferograms[J]. Eurasip Journal on Advancesin Signal Processing,2008
[77] HERNANDEZ O, OLVERA E. Noise Cancellation On ECG And Heart Rate SignalsUsing The Undecimated Wavelet Transform[C]. Proceedings International Conference oneHealth, Telemedicine, and Social Medicine. Cancun, Mexico,2009:145-150
[78] DONOHO D. L., JOHNSTONE I M. Ideal Spatial Adaptation Via Wavelet Shrinkage[J].Biometrika,1994,81(3):425-455.
[79] DONOHO D. L. De-Noising By Soft-Thresholding[J]. IEEE Transactions onInformation Theory,1995,41(3):613-627
[80] DONOHO D. L., JOHNSTONE I. M. Adapting To Unknown Smoothness Via WaveletShrinkage[J]. Journal of the American Statistic Association,1995,90(432):1200-1224
[81] SELESNICK I. W. The Double Density DWT. In A. Petrosian and F. G. Meyer, editors,Wavelets in signal and image analysis: from theory to practice[Z],2001.http://taco.poly.edu/selesi/.
[82]杨恢先,王绪四,谢鹏鹤,冷爱莲,彭友.改进阈值与尺度间相关的小波红外图像去噪[J].自动化学报,2011,37(10):1167-1174.
[83]邹海林,隋亚莉,徐俊艳,宁书年.基于多小波变换的GPR图像去噪方法研究[J].系统仿真学报,2005,17(4):855-858,862.
[84] HONGYE GAO, Bruce A. G. Waveshrink and Semisoft Shrinkage[R]. StaSci ResearchReport No.39,1995.
[85]罗志增,张清菊,蒋静坪.表面肌电信号的小波消噪改进算法[J].浙江大学学(工学版).2007,41(2):213-220.
[86]李阳,田彦涛,陈万忠.基于FFT盲辨识的肌电信号建模及模式识别[J].自动化学报,2011,38(1):128-134
[87] YANG LI, YANTAO TIAN, WANZHONG CHEN. Multi-pattern Recognition of sEMGBased on Improved BP Neural Network Algorithm[C]. Proceedings of the29th ChineseControl Conference. Beijing, China.2010:2867-2872.
[88] GUANGHAN XU, HUI LIU, TONG LANG. Least Squares Approach to Blind ChannelIdentification[J]. IEEE Transactions on Signal Processing,1995,43(12):2982-2993
[89] ERWEI BAI. A Blind Approach to The Hammerstein Model Identification[J]. IEEETransactions on Signal Processing.2002,50(7):1610-1619
[90] NARASIMHAN S. V., HAZARATHAIAH M., GIRIDHAR P. Channel BlindIdentification Based on Cyclostationarity And Group Delay[J]. IEEE Transactions onSignal Processing.2005,85(7):1275-1286
[91] FANG J., LEYMANB A. R., CHEW Y. H. Some Further Results on Blind Identificationof MIMO FIR Channels via Second-Order Statistics[J]. IEEE Transactions on SignalProcessing.2007,87(6):1434-1447
[92]徐小平,钱富才,王峰.一种辨识Wiener-Hammerstein模型的新方法[J].控制与决策,2008,23(8):929-934
[93] FARINA D., MERLETTI R. Comparison of Algorithms for Estimation of EMGVariables During Voluntary Inometric Contractions[J]. Electromyogr. Kinesiol.2000,10:337-349.
[94] UCHIDA N. U., HIRAIWA A., SONEHARA N., et al. EMG Pattern Recognition byNeural Networks for Multi Fingers Control[C]. The Annual International Conference of theEngineering in Medicine and Biology Society. Paris, France,1992,14:1016-1018.
[95] FARRY K. A., WALKER I. D.. Myoeletric Teleoperation of a Complex RoboticHand[C]. Proceeding of the1993IEEE International Conference on Robotics andAutomation. Atlanta, GA,1993:502-509.
[96] ZHIZENG LUO, YUTAO JIA. Multi-pattern recognition by using surfaceelectromyography’s power spectra and BP neural networks [J]. J.Huanzhong Univ. of Sci.&Tech.(Nature Science Edition),2006,34(7):63-66.
[97] YANG LI, YANTAO TIAN, XIAOJING SHANG, WANZHONG CHEN. Modeling andClassification of sEMG Based on Blind Identification Theory[C].8th InternationalSymposium on Neural Networks. Guilin, China,2011:340-347
[98] XIAOJING SHANG, YANTAO TIAN, YANG LI. Modeling and Classification of sEMGBased on Instrumental Variable Identification[C].8th International Symposium on NeuralNetworks. Guilin, China,2011:331-339
[99] YITENG HUANG, JACOB B., JINGDONG CHEN. Using the Pearson CorrelationCoefficient to Develop An Optimally Weighted Cross Relation Based SIMO IdentificationAlgorithm[C]. Acoustics, Speech and Signal Processing. Taipei, China,2009:1520-6149
[100] MITCHELL T. M.,曾华军,张银奎.机器学习[M].北京：机械工业出版社.2003
[101] BENNETT K. P., DEMIRIZ A. Semi-Supervised Support Vector Machine[M]. NIPS.1999
[102] CHAPELLE O., ZIEN A. Semi-Supervised Classification by Low DensitySeperation[C].10thInt. Workshop on AI and Stat.2005.
[103] BLUM A., CHAWLA S. Learning From Labeled and Unlabeled Data Using GraphMincuts[C]. ICML.2001.
[104] XIAOJIN ZHU, GHAHRAMANI Z., LAFFERTY J. Semi-Supervised Learning UsingGaussian Fields and Harmonic Functions. ICML.2003.
[105] ZHOU D., BOUSQUET O., LAL T., WESTON J., SCHOLKOPF B. Learning WithLocal and Global Consistency[C]. COLT.2004
[106] BELKIN M., MATVEEVA I., NIYOGI P. Regularization and Semisupervised Learningon Large Graphs[C]. COLT.2004
[107] BENNETT K. P., DEMIRIZ A., MACLIN R. Exploiting Unlabeled Data in EnsembleMethods[C]. KDD.2002.
[108] KE CHEN, SHIHAI WANG. Regularized Boost for Semi-Supervised Learning. NIPS.2007.
[109] DIETTERICH T. G., BAKIRI G. Solving Multiclass Learning Problems Via ErrorCorrecting Output Codes[J]. AI Res.1995,2:263-286.
[110] RONG JIN, JIAN ZHANG. Multi-Class Learning by Smoothed Boosting[J]. Mach.Learn.2007,67(3):207-227.
[111] SCHOLKOPF B., SMOLA A. J. Learning with Kernels: Support Vector Machines,Regularization, Optimization, and Beyond[M]. USA: MIT Press.2001
[112] ZADROZNY B., ELKAN C. Transforming Classifier Scores Into Accurate MulticlassProbability Estimates[C]. KDD.2002
[113] EIBL G., PFEIFFER K. P. Multiclass Boosting for Weak Classifiers[J]. Mach. Learn.Res.2005,6:189-210.
[114] NG LI. Multiclass Boosting with Repartitioning[C]. ICML, Pittsburgh, USA,2006:569-576
[115] DALCHE B. F., GRANDVALET Y., AMBROISE C. Semi-Supervised Marginboost[C].NIPS.2002.
[116] XU ZHU. Semi-Supervised Learning Literature Survey[R]. Technical Report1530,Computer Science, University of Wisconsin-Madison.2005.
[117] JOACHIMS T. Transductive Inference for Text Classification Using Support VectorMachines[C]. ICML.1999.
[118] BIE T. D., CRISTIANINI N. Convex Methods for Transduction[C]. NIPS.2004.
[119] XU L., SCHUURMANS D. Unsupervised and Semi-Supervised Multi-Class Supportvector Machines[C]. AAAI.2005.
[120] FREUND Y., SCHAPIRE R. E. Experiments with A New Boosting Algorithm[C].ICML.1996.
[121] HIGHAM N. J. Matrix Nearness Problems And Applications[M]. Gover, M.J.C.,Barnett S., eds.: Applications of Matrix Theory, UK: Oxford University Press.1989.
[122] OPITZ D., MACLIN R. Popular Ensemble Methods: An Empirical Study[J]. AI Res.1999,11:169-198.
[123] SHI J., MALIK J. Normalized Cuts and Image Segmentation[J]. IEEE Transactions onPattern Analysisi and Machine Intelligence.2000,22(8):888-905.
[124]王国强.虚拟样机技术及其在ADAMS上的实践[M].西安：西北工业大学出版社.2002.
[125]王洪亮,赵丽萍,陈富民.机械设计软件应用[M].西安：西安交通大学出版社.2007.
[126]郭卫东.虚拟样机技术与Adams应用实例教程[M].北京：北京航空航天大学出版社.2008.
[127]韩京清.非线性状态误差反馈控制律——NLSEF[J].控制与决策.1995,10(3):221-231.
[128]韩京清.从PID技术到“自抗扰控制”技术[J].控制工程.2002,9(3):13-18.
[129]陈慧琴.基于自抗扰控制器的永磁同步电动机交流调速系统的研究[D].天津大学硕士论文.2004
[130]韩京清,张荣.二阶扩张状态观测器的误差分析[J].系统科学与数学.1999,19(4):465-471.
[131]黄一,韩京清.非线性连续二阶扩张状态观测器的分析与设计[J].科学通报.2000,45(13):1373-1379.
[132]李海生,朱学峰.自抗扰控制器参数整定与优化方法研究[J].控制工程.2004,11(5):419-423.
[133]韩京清.自抗扰控制技术——估计补偿不确定因素的控制技术[M].北京：国防工业出版社.2009
[134] WEIQIANG PAN, HAIRONG XIAO, CHANGSHUN WANG. Design of Ship CourseController Based on Optimal Active Disturbance Rejection Technique[C]. InternationalConference on Intelligent System Design and Engineering Application, ShangDong, China,2010:582-585.
[135] YONGJUN LI, YUANFU LU, CHAO Hu. An Active Disturbance Rejection ControlTechnique for the Turntable System with Motor Moment Fluctuation[C].8thIEEEInternational Conference on Control and Automation, ShenZhen, China,2010:80-85.
[136]朱晞.人体解剖学[M].浙江：浙江大学出版社，2003.
[2]EDUARDO MANGIERI et al. A Novel Analogue Circuit for Controlling ProstheticHands[C]. IEEE Biomedical Circuits and Systems Conference, USA,2008:81-84.
[3]MERLETTI R., RAINOLDI A., FARIAN D. Surface electromyography for noninvasivecharacterization of muscle[J]. Exercise and Sport Sciences Reviews.2001,29:20-25.
[4] BICCHI A. Hand for Dexterous Manipulation and Robust Grasping: A Difficult RoadToward Simplicity[J]. IEEE Trans. Robot. Autom.2000,16(6):652-662.
[5] BIAGIOTTI L., LOTTI F., MELCHIORRI C., VASSURA G.. How Far Is the HumanHand? A Review on Anthroponorphic Robotic End-effectors. Italy: University ofBologna.[2009-04-05].http://www-lar.deis.upioe.it/woda/data/deis-lar-publications/3cbd.Document.pdf
[6] DASHY R., VEI C., LABIANC M.. The Design and Development of a GlovelessEndoskeleton of Prosthetic Hand[J]. Journal of Rehabilitation Research and Development.1998,35(4):388-395.
[7] DECHEV N., LEGHORN W. L., NEUMANN S.. Multiple Finger Passive AdaptiveGrasp Prosthetic Hand[J]. Mechanism and Machine Theory.2001,36:1157-1173.
[8] KYBERD P. J., LIGHT C. M., CHAPPEL P. H., et al. The Design of AnthropomorphicProsthetic Hands: A Study of the Southampton Hand[J]. Robotica.2001,19(6):593-600.
[9] LIGHT C. M., CHAPPELL P. H.. Development of a lightweight and adaptablemultiple-axis hand prosthesis[J]. Medical Engineering and Physics.2000,22(10):679-684.
[10] KYBERD P. J.. The intelligent hand[J]. IEEE Review.2000,9:31-35.
[11] ZHIXIU HAO, YONGSHENG SHEN. Force self-adaptive control for prehensors[J]. JTsinghua Univ (Sci&Tech),1999,39:92-99.
[12] HIRAIWA A., SHIMOHARA K., TOKUNAGA Y.. EMG Pattern Analysis andClassification by Neural Network[C]. IEEE International Conference on Systems, Man andCybernetics, Cambridge, MA, USA,1989:1113-1115.
[13] LOTTI F., TIEZZI P., MELCHIORRI C.. Development of UB Hand3[C]. Proceedingsof the2005IEEE International Conference on Robotics and Automation, Barcelona, Spain,2005:4499-4503.
[14] ZOLLO L., ROCCELLA S., GUGLIELMELLI E., et al. Biomechatronic Design andControl of an Anthropomorphic Artificial Hand for Prosthetic and Robotic Applications[J].IEEE/ASME Transactions on Mechatronics.2007,12(4):418-429.
[15] PYLATIUK C., MOUNIER S., KARGOV A., et al. Progress in the Development of aMultifunctional Hand Prosthesis[C]. The26thIEEE Annual International Conference onEngineering in Medicine and Biology Society, San Francisco, USA.2004:4260-4263.
[16] The i-limb system[Z].[2010-5-27]. www.touchbionics.com
[17] Otto Bock Michelangelo hand[Z].[2009-03-27]. http://www.swisswuff.ch/tech/?p=145.
[18] HE. HUANG, PING. ZHOU, et al. An Analysis of EMG Electrode Configuration forTargeted Muscle Reinnervation Based Neural Machine Interface[J]. IEEE Trans. NeuralSystems and Rehabilitation Engineering.2008,16(1):37-46.
[19] KUIKEN T. A., LEE G., LOCK B. A., et al. Targeted Muscle Reinnervation forReal-time Myoelectric Control of Multifuction Artificial Arms[J]. Journal of the AmericanMedical Association.2009,301(6):619-628.
[20] DALLEY S. A., WISTE T. E., WITHROW T. J., et al. Design of a MultifuctionalAnthropomorphic Prosthetic Hand With Extrinsic Actuation[J]. IEEE Transactions onMechatronics.2009,14(6):699-706.
[21]杨大鹏.特定心理作业多通道脑电信号识别系统研究[D].哈尔滨工业大学硕士学位论文.2006.
[22] ENGEL A. K., MOLL C. K. E., FRIED I., et al. Invasive Recordings from the HumanBrain: Clinical Insights and Beyond[J]. NATURE.2005,6:35-47.
[23] MEEL V., SAGI P., SPALDING C. M., et al. Cortical Control of A Prosthetic Arm forSelf-Feeding[J]. NATURE.2008,452(19):1098-1101.
[24] ZARDOSHTI K. M., WHEELER B. C., BADIE K., HASHEMI R. M. EMG FeatureSelection for Movement Control of a Cybernetic Arm[J]. Cybernetics and Systems.1995,26(2):189-210.
[25] BOOSTANI R., MORADI M. H. Evaluation of the Forearm EMG Signal Feature for theControl of a Prosthetic Hand[J]. Physiological Measurement.2003,24(2):309-319.
[26] KANG W. J., CHENG C. K., LAI J. S., SHIU J. R., KUO T. S. A Comparative Analysisof Various EMG Pattern Recognition Methods[J]. Medical Engineering and Physics.1996,18(5):390-395.
[27] KIM W. S., YOON Y. Z., BAE J. H., SOH K. S. Nonlinear Characteristics of Heart RateTime Series: Influence of Three Recumbent Positions in Patients With Mild or SevereCoronary Artery Disease[J]. Physiological Measurement.2005,26(4):517-529.
[28] INBAR G. F., ALLINE J., PAISS O., KRANZ H. Monitoring Surface EMG SpectralChanges By the Zero Crossing Rate[J]. Medical and Biological Engineering andComputing.1986,24(1):10-18.
[29] MERLETTI R., LOCONTE L. R., ORIZIO C. Indices of Muscle Fatigue[J]. Journal ofElectromyography and Kinesiology.1991,1(1):20-33.
[30] FARIAN D, MERLETTO R. Comparison of Algorithms for Estimation of EMGVariables During Voluntary Isometric Contractions[J]. Journal of Electromyography andKinesiology.2000,10(5):337-349.
[31] UCHIDA N., HIRAIWA A., SONEHARA N., SHIMOHARAL K.. EMG patternrecognition by neural networks for multi finger control[C]. Proc. Ann. Eng. Med. Biol.Soc.,1992:1016–1018.
[32] RONAGER J., CHRISTENSEN H. Power Spectrum Analysis of the EMG Pattern inNormal and Diseased Muscles[J]. Journal of the Neurologicla Sciences.1989,94(1-3):283-294.
[33] BILODIAU M., CINCERA M., GERVAIS S., et al. Changes in the ElectromyographicSpectrum Power Distribution Caused by a Progressive Increase in the Force Level[J].European Journal of Applied Physiology and Occupational Physiology.1995,71(2-3):113-123.
[34] ECKARDT L., HARZER W. Facial Structure and Functional Findings in Patients withProgressive Muscular Dystrophy(Duchenne)[J]. American Journal of Orthodontics andDentofacial orthopedics: Official Publication of the American Association of Orthodontists,Its Constituent Societies, and the American Board of Orthodontics.1996,110(2):185-190.
[35] FARINA D. Interpretation of the Surface Eletromyogram in Dynamic Contractions[J].Exercise and Sport Sciences Reviews.2006,34(3):121-127.
[36]王飞,罗志增.基于功率谱的表面EMG信号运动模式识别[J].杭州电子科技大学学报.2005,25(2):37—40.
[37] DAVIES M. R., REISMAN S. S. Time Frequency Analysis of the ElectromyogramDuring Fatigue[C]. Bioengineering Proceedings of the Northeast Conference, Springfield,MA, USA.1994
[38] JANG G. C., CHENG C. K., LAI J. S., KUO T. S. Using Time-Frequency AnalysisTechnique in the Classification of Surface EMG Signals[C]. IEEE Engineering inMedicine and Biology Proceedings, Baltimore, MD, USA.1994.
[39] BONATO P., GAGLIATI G., KNAFLITZ M. Analysis of Myoelectric Signals RecordedDuring Dynamic Contractions[J]. IEEE Engineering in Medicine and Biology Magazine.1996,15(6):102-111.
[40] BECK T. W., HOUSH T. J., JOHNSON G. O., WEIR J. P., et al. Comparison of Fourierand Wavelet Transform Procedures for Examining the Mechanomyographic andElectromyographic Frequency Domain Response During Fatiguing Isokinetic MuscleActions of the Biceps Brachii[J]. Journal of Electromyography and Kinesiology.2005,15(2):190-199.
[41] MERLO A., FARINA D., MERLETTI R. A Fast and Reliable Technique for MuscleActivity Detection From Surface EMG Signals[J]. IEEE Transactions on BiomedicalEngineering.2003,50(3):316-323.
[42] KARLSSON S., YU J., AKAY M. Time-Frequency Analysis of Myoelectric SignalsDuring Dynamic Contractions: A Comparative Study[J]. IEEE Transactions on BiomeicalEngineering.2000,47(2):228-238.
[43] HOBO XIE, ZHIZHONG Wang. Mean Frequency Derived Via Hilbert-HuangTransform with Application to Fatigue EMG Signal Analysis[J]. Computer Methods andPrograms in Biomedicine.2006,82(2):114-120.
[44] CHAN F. H. Y., YANG Y. S., LAM F. K., ZHANG Y. T., PARKER P. A.. Fuzzy EMGClassification for Prosthesis Control[J]. IEEE Trans. Rehabil. Eng.,2000,8(3):305-311.
[45] PINGAO MEI, ZHIZENG LUO. sEMG Disposal Based on Wavelet Packet Analysis andElman Neural Network[J]. Mechanical&Electrical Engineering Magazine,2008,25:7-10.
[46] YIHUNG LIU, HANPANG HUANG, CHANGHSIN WENG. Recognition ofElectromyographic Signals Using Cascaded Kernel Learning Machine[J]. IEEE/ASMETransactions on Mechatronics,12:253-264.
[47] ZHIHONG LIU, ZHIZENG LUO. Hand Motion Pattern Classifier Based on EMGUsing Wavelet Packet Transform and LVQ Neural Networks[C]. Proceedings of2008IEEE International Symposium on IT in Medicine and Education, China,2008:28-32.
[48] JINGDONG ZHAO, JIANG LI, MINGHE JIN, HONG LIU, HEGAO CAI. ProstheticHand Control Based on EMG Signal[J]. Machinary&Electron,2006,8:54-56.
[49] HAIHONG ZHANG, LIYU CAI, ZHIZHONG WANG. A Pi-Sigma neural networkbased electromyograph signal identification method[J]. Chinese Journal of RehabilitationMedicine.2000,15:34-37.
[50] XIANG CHEN, XU ZHANG, ZHANGYAN ZHAO, et al. Multiple Hand GestureRecognition based on Surface EMG Signal[C].1stInternatinal Conference onBioinformatics and Biomedical Engineering, China,2007:506—509.
[51] ZHIZENG LUO, FEI WANG. Electromyography Movement Pattern Recognition Basedon the Wavelet Eigenvalues and Pi-sigma Network[J]. Journal of Test And MeasurementTechnology,2006,20:344-348.
[52]李强.表面肌电信号的运动单位动作电位检测[D].中国科学技术大学博士论文.2008.
[53] CLANCY E. A., MORIN E. L., MERLETTI R. Sampling, Noise-Reduction andAmplitude estimation Issues in Surface Electromyography[J]. Journal ofElectromyography and Kinesiology.2002,12(1):1-16
[54] REAZ M. B. I., HUSSAIN M. S., MOHDYASIN F. Techniques of EMG Signal Analysis:Detection, Processing, Classification and Applications[J]. Biological Procedures Online.2006:11-35.
[55] PARKER. Signal Processing for Multistate Myoeletric Channel[J]. Proc. IEEE.1997,65(5):662-670.
[56] PARKER. Myoeletric Control of Prosthsis[J]. CRC Critical Review BiomedicalEngineering.1986,13(4):103-107.
[57] SHWEDYK. A Nonstationary Model for the Electromyogram[J]. IEEE Trans. BME-24.1997:417-429.
[58] CHEUNG J，HALBERTSMA J P，VELDHUIZEN A G. A Preliminary Study onElectromyographic Analysis of the Para Spinal Musculature in Idiopathic Scoliosis[J]. EurSpine J，2005，14(2):130-137.
[59] SHERIF M. H. Effects of Load on Myoeletric Signals: The ARIMA Representation[J].IEEE Trans. Biomed. Eng.1981,28:411-416.
[60] Brain-controlled'robo-arm'hope, BBC NEWS UK EDITION,Friday,18February,2005.
[61] KATOH R., HIROSHI Y., TAMIO A. Competitive Learning Method for RobustEMG-to-Motion Classifier[C]. IAS-9,2006:7-9.
[62] GRAUPE D. Multifunctional Prosthesis and Orthosis Control ViamicrocomputerIdentification of Temporal Pattern Differneces in Singlesite Myoeletric Signals[J]. Biomed.Eng.1982,4:17-22.
[63] JINGDONG ZHAO, ZONGWU XIE, HONG LIU, et al. Levenberg-Marquardt BasedNeural Network Control for a Five-fingered Prosthetic Hand[C]. IEEE Int. Conf.Roboticsand Automation, China,2005:4493–4498.
[64]汤小芙.临床肌电图学[M].北京：北京医科大学中国协和医科大学联合出版社.1995
[65]蔡立羽,王志中,刘瑜.基于盲辨识理论的双通道肌电信号建模与分类[J].上海交通大学学报.2000,34(11):1471-1474.
[66] DELUCA C. Physiology and Mathematics of Myoeletrics Signals[J]. IEEE Trans.Biomed. Eng.1979,26(6):313-325.
[67] MIN LEI, ZHIZHONG WANG, ZHENGJIN FENG. Detecting Nonlinearity of ActionSurface EMG Signal[J]. Physics Letters A.2001,290:297-303.
[68] BILODIAU M. Variations in the Shape of the EMG Power Specturm During anIncreasing Force Level[C]. Annual International Conference of the IEEE Engineering inMedicine&Biology Society-Proceedings.1993,15:1205-1206.
[69]何乐生.基于肌电信号的人机接口技术的研究[D].东南大学博士论文.2006.
[70] OSKOEI M., HU H. Myoelectric Control Systems-A Survey[J]. Biomedical SignalProceeding and Control.2007,2(4):275-294.
[71] SREE P S J, KUMAR P, SIDDAVATAM R, VERMA R. Slat-And-Pepper NoiseRemoval By Adaptive Median-Based Lifting Filter Using Second-Generation Wavelets[J].Signal, Image and Video Processing,2011:1-8(Article in Press)
[72] WEIXIA ZHAN, JIWEN TAN, YAN WEN. De-noising of Damage Signal in Wire RopeBased on Adaptive Lifting Wavelet[J]. Manufacturing Engineering and Automation,2010,139-141:2556-2560
[73] NONOMURA K, ONO M, ZHOU L B, SHIMIZU J, OJIMA H. Design of Digital FiltersFor Si Wafer Surface Profile Measurement-Noise Reduction By Lifting Scheme WaveletTransform-Noise[J]. Advances in Abrasive Technology XIII,2010,126-128:732-737
[74] INOUE H, MIYAZAKI A. Noise Reduction Method For ECG Signals Using The DyadicWavelet Transform[J]. IEICE Transactions on Fundamentals of ElectronicsCommunications and Computer Sciences,1998, E81-A(6):1001-1007
[75]祝海龙，郭天佑，屈梁生.基于二进小波变换和软阈值改进的信号消噪[J].自动化学报，2004,30(2):199-206
[76] SETHU V, AMBIKAIRAJAH E, Ge L. Selective Weighting Of Undecimated WaveletCoefficients For Noise Reduction In SAR Interferograms[J]. Eurasip Journal on Advancesin Signal Processing,2008
[77] HERNANDEZ O, OLVERA E. Noise Cancellation On ECG And Heart Rate SignalsUsing The Undecimated Wavelet Transform[C]. Proceedings International Conference oneHealth, Telemedicine, and Social Medicine. Cancun, Mexico,2009:145-150
[78] DONOHO D. L., JOHNSTONE I M. Ideal Spatial Adaptation Via Wavelet Shrinkage[J].Biometrika,1994,81(3):425-455.
[79] DONOHO D. L. De-Noising By Soft-Thresholding[J]. IEEE Transactions onInformation Theory,1995,41(3):613-627
[80] DONOHO D. L., JOHNSTONE I. M. Adapting To Unknown Smoothness Via WaveletShrinkage[J]. Journal of the American Statistic Association,1995,90(432):1200-1224
[81] SELESNICK I. W. The Double Density DWT. In A. Petrosian and F. G. Meyer, editors,Wavelets in signal and image analysis: from theory to practice[Z],2001.http://taco.poly.edu/selesi/.
[82]杨恢先,王绪四,谢鹏鹤,冷爱莲,彭友.改进阈值与尺度间相关的小波红外图像去噪[J].自动化学报,2011,37(10):1167-1174.
[83]邹海林,隋亚莉,徐俊艳,宁书年.基于多小波变换的GPR图像去噪方法研究[J].系统仿真学报,2005,17(4):855-858,862.
[84] HONGYE GAO, Bruce A. G. Waveshrink and Semisoft Shrinkage[R]. StaSci ResearchReport No.39,1995.
[85]罗志增,张清菊,蒋静坪.表面肌电信号的小波消噪改进算法[J].浙江大学学(工学版).2007,41(2):213-220.
[86]李阳,田彦涛,陈万忠.基于FFT盲辨识的肌电信号建模及模式识别[J].自动化学报,2011,38(1):128-134
[87] YANG LI, YANTAO TIAN, WANZHONG CHEN. Multi-pattern Recognition of sEMGBased on Improved BP Neural Network Algorithm[C]. Proceedings of the29th ChineseControl Conference. Beijing, China.2010:2867-2872.
[88] GUANGHAN XU, HUI LIU, TONG LANG. Least Squares Approach to Blind ChannelIdentification[J]. IEEE Transactions on Signal Processing,1995,43(12):2982-2993
[89] ERWEI BAI. A Blind Approach to The Hammerstein Model Identification[J]. IEEETransactions on Signal Processing.2002,50(7):1610-1619
[90] NARASIMHAN S. V., HAZARATHAIAH M., GIRIDHAR P. Channel BlindIdentification Based on Cyclostationarity And Group Delay[J]. IEEE Transactions onSignal Processing.2005,85(7):1275-1286
[91] FANG J., LEYMANB A. R., CHEW Y. H. Some Further Results on Blind Identificationof MIMO FIR Channels via Second-Order Statistics[J]. IEEE Transactions on SignalProcessing.2007,87(6):1434-1447
[92]徐小平,钱富才,王峰.一种辨识Wiener-Hammerstein模型的新方法[J].控制与决策,2008,23(8):929-934
[93] FARINA D., MERLETTI R. Comparison of Algorithms for Estimation of EMGVariables During Voluntary Inometric Contractions[J]. Electromyogr. Kinesiol.2000,10:337-349.
[94] UCHIDA N. U., HIRAIWA A., SONEHARA N., et al. EMG Pattern Recognition byNeural Networks for Multi Fingers Control[C]. The Annual International Conference of theEngineering in Medicine and Biology Society. Paris, France,1992,14:1016-1018.
[95] FARRY K. A., WALKER I. D.. Myoeletric Teleoperation of a Complex RoboticHand[C]. Proceeding of the1993IEEE International Conference on Robotics andAutomation. Atlanta, GA,1993:502-509.
[96] ZHIZENG LUO, YUTAO JIA. Multi-pattern recognition by using surfaceelectromyography’s power spectra and BP neural networks [J]. J.Huanzhong Univ. of Sci.&Tech.(Nature Science Edition),2006,34(7):63-66.
[97] YANG LI, YANTAO TIAN, XIAOJING SHANG, WANZHONG CHEN. Modeling andClassification of sEMG Based on Blind Identification Theory[C].8th InternationalSymposium on Neural Networks. Guilin, China,2011:340-347
[98] XIAOJING SHANG, YANTAO TIAN, YANG LI. Modeling and Classification of sEMGBased on Instrumental Variable Identification[C].8th International Symposium on NeuralNetworks. Guilin, China,2011:331-339
[99] YITENG HUANG, JACOB B., JINGDONG CHEN. Using the Pearson CorrelationCoefficient to Develop An Optimally Weighted Cross Relation Based SIMO IdentificationAlgorithm[C]. Acoustics, Speech and Signal Processing. Taipei, China,2009:1520-6149
[100] MITCHELL T. M.,曾华军,张银奎.机器学习[M].北京：机械工业出版社.2003
[101] BENNETT K. P., DEMIRIZ A. Semi-Supervised Support Vector Machine[M]. NIPS.1999
[102] CHAPELLE O., ZIEN A. Semi-Supervised Classification by Low DensitySeperation[C].10thInt. Workshop on AI and Stat.2005.
[103] BLUM A., CHAWLA S. Learning From Labeled and Unlabeled Data Using GraphMincuts[C]. ICML.2001.
[104] XIAOJIN ZHU, GHAHRAMANI Z., LAFFERTY J. Semi-Supervised Learning UsingGaussian Fields and Harmonic Functions. ICML.2003.
[105] ZHOU D., BOUSQUET O., LAL T., WESTON J., SCHOLKOPF B. Learning WithLocal and Global Consistency[C]. COLT.2004
[106] BELKIN M., MATVEEVA I., NIYOGI P. Regularization and Semisupervised Learningon Large Graphs[C]. COLT.2004
[107] BENNETT K. P., DEMIRIZ A., MACLIN R. Exploiting Unlabeled Data in EnsembleMethods[C]. KDD.2002.
[108] KE CHEN, SHIHAI WANG. Regularized Boost for Semi-Supervised Learning. NIPS.2007.
[109] DIETTERICH T. G., BAKIRI G. Solving Multiclass Learning Problems Via ErrorCorrecting Output Codes[J]. AI Res.1995,2:263-286.
[110] RONG JIN, JIAN ZHANG. Multi-Class Learning by Smoothed Boosting[J]. Mach.Learn.2007,67(3):207-227.
[111] SCHOLKOPF B., SMOLA A. J. Learning with Kernels: Support Vector Machines,Regularization, Optimization, and Beyond[M]. USA: MIT Press.2001
[112] ZADROZNY B., ELKAN C. Transforming Classifier Scores Into Accurate MulticlassProbability Estimates[C]. KDD.2002
[113] EIBL G., PFEIFFER K. P. Multiclass Boosting for Weak Classifiers[J]. Mach. Learn.Res.2005,6:189-210.
[114] NG LI. Multiclass Boosting with Repartitioning[C]. ICML, Pittsburgh, USA,2006:569-576
[115] DALCHE B. F., GRANDVALET Y., AMBROISE C. Semi-Supervised Marginboost[C].NIPS.2002.
[116] XU ZHU. Semi-Supervised Learning Literature Survey[R]. Technical Report1530,Computer Science, University of Wisconsin-Madison.2005.
[117] JOACHIMS T. Transductive Inference for Text Classification Using Support VectorMachines[C]. ICML.1999.
[118] BIE T. D., CRISTIANINI N. Convex Methods for Transduction[C]. NIPS.2004.
[119] XU L., SCHUURMANS D. Unsupervised and Semi-Supervised Multi-Class Supportvector Machines[C]. AAAI.2005.
[120] FREUND Y., SCHAPIRE R. E. Experiments with A New Boosting Algorithm[C].ICML.1996.
[121] HIGHAM N. J. Matrix Nearness Problems And Applications[M]. Gover, M.J.C.,Barnett S., eds.: Applications of Matrix Theory, UK: Oxford University Press.1989.
[122] OPITZ D., MACLIN R. Popular Ensemble Methods: An Empirical Study[J]. AI Res.1999,11:169-198.
[123] SHI J., MALIK J. Normalized Cuts and Image Segmentation[J]. IEEE Transactions onPattern Analysisi and Machine Intelligence.2000,22(8):888-905.
[124]王国强.虚拟样机技术及其在ADAMS上的实践[M].西安：西北工业大学出版社.2002.
[125]王洪亮,赵丽萍,陈富民.机械设计软件应用[M].西安：西安交通大学出版社.2007.
[126]郭卫东.虚拟样机技术与Adams应用实例教程[M].北京：北京航空航天大学出版社.2008.
[127]韩京清.非线性状态误差反馈控制律——NLSEF[J].控制与决策.1995,10(3):221-231.
[128]韩京清.从PID技术到“自抗扰控制”技术[J].控制工程.2002,9(3):13-18.
[129]陈慧琴.基于自抗扰控制器的永磁同步电动机交流调速系统的研究[D].天津大学硕士论文.2004
[130]韩京清,张荣.二阶扩张状态观测器的误差分析[J].系统科学与数学.1999,19(4):465-471.
[131]黄一,韩京清.非线性连续二阶扩张状态观测器的分析与设计[J].科学通报.2000,45(13):1373-1379.
[132]李海生,朱学峰.自抗扰控制器参数整定与优化方法研究[J].控制工程.2004,11(5):419-423.
[133]韩京清.自抗扰控制技术——估计补偿不确定因素的控制技术[M].北京：国防工业出版社.2009
[134] WEIQIANG PAN, HAIRONG XIAO, CHANGSHUN WANG. Design of Ship CourseController Based on Optimal Active Disturbance Rejection Technique[C]. InternationalConference on Intelligent System Design and Engineering Application, ShangDong, China,2010:582-585.
[135] YONGJUN LI, YUANFU LU, CHAO Hu. An Active Disturbance Rejection ControlTechnique for the Turntable System with Motor Moment Fluctuation[C].8thIEEEInternational Conference on Control and Automation, ShenZhen, China,2010:80-85.
[136]朱晞.人体解剖学[M].浙江：浙江大学出版社，2003.