用于设备操控件布局设计的人体静态姿势评价准则

根据人体肌肉硬度与肌肉舒适度具有因果关系的生理学特性,进行了由一组志愿者参与的、旨在研究操控件不同布局设计下人体静态姿势对肌肉硬度影响的试验。利用肌肉硬度仪测量了一组受试者在不同静态操作姿势下人体各相关肌肉硬度的变化情况,得到不同姿势下肌肉硬度的变化规律。通过对试验测得的肌肉硬度数据进行归一化处理并依据相关肌肉横断面积进行加权计算,给出肌肉舒适指数的概念;针对人体上身2自由度、3自由度关节动作情况,分别给出综合肌肉舒适指数的计算方法;最终提出了基于肌肉硬度试验数据的人体舒适度评价准则(Assessment system based on muscle hardness experiments,ASMH)。通过与工程中广泛采用的快速上肢评价准则(Rapid upper limb assessment,RULA)对比可知,此评价准则的准确度和适用性有所提高,评分值范围增加为1至12,评价等级增加至六个等级,更适合于操控件布局设计的人机工程学优化设计。     

修井机司钻作业生物力学仿真与姿态舒适度评价

为更加准确地评价修井机司钻人员作业姿态舒适度,在生物力学分析软件AMS中构建了第50百分位的中国成年男子人体数字模型,并对司钻作业主要动作进行了逆向动力学仿真。以该动作相关肌群的肌肉激活度为评价指标,以坐高、靠背倾角、脚踏位置为变量,运用MATLAB软件对仿真结果进行拟合处理,分别构建了司钻作业姿态舒适度定量评价模型,为修井机人机工程评价及操作舱人机布置改进设计提供了可靠依据。     

一种肌电假手的自适应模糊控制方法

针对肌电假手的力控制问题,提出了一种基于肌电信号自适应学习的动作识别方法,同时结合模糊神经网络PID控制算法实现肌电信号对假手的控制.设计的肌电信号自适应学习动作识别方法在时域内进行,减小运算复杂度和计算量的同时保证了动作识别精度.采用模糊神经网络PID算法设计了假手握力控制器,在没有位置传感器的情况下保证了假手握力的控制精度.进行了肌电信号动作识别跟踪实验、假手握力跟踪实验、肌电信号控制假手抓取实验,实验结果证明了肌电信号自适应学习动作识别方法和模糊神经网络PID握力控制方法在肌电假手控制中的有效性.     

基于脑肌电反馈的虚拟康复训练系统设计

针对虚拟康复系统在个体适应性、安全可行性和主动参与性等方面的研究需求,提出一种基于脑肌电反馈的虚拟康复系统。采集执行不同手势动作对应的脑电信号及表面肌电信号,提取不同肌肉模块的肌电特征送入支持向量机模型进行运动意图识别;提取脑电和肌电疲劳特征,并提出一种萤火虫-模糊神经网络算法,通过脑肌电疲劳特征实时优化调节虚拟场景的控制参数。最后,搭建包含虚拟场景及反馈控制策略的虚拟康复系统,并针对上肢肘关节屈伸、肩关节前屈后伸动作进行康复训练实验,基于肌电特征模式识别结果实现对虚拟场景及目标的控制,基于脑肌电疲劳特征优化调整场景控制参数,通过虚拟系统康复实验验证系统的有效性。     

肌电调制结合迭代学习控制的足下垂 FES 系统

足下垂是指由于神经控制功能障碍导致踝关节无法产生背屈以致足尖上抬不完全或不能的现象。功能性电刺激(FES)作为纠正足下垂步态的治疗方法,通过低频脉冲刺激胫骨前肌引起肌肉收缩,使踝关节产生背屈动作,达到矫正足下垂的目的。本文提出了基于肌电(EMG)调制和迭代学习控制(ILC)的FES输出强度调制方法,利用小腿角速度信号通过动态BP神经网络预测健康步态胫骨前肌肌电信号,以脚尖俯仰角作为反馈信号通过ILC输出参考肌电信号,与神经网络预测的肌电信号加权平均得到修正后的肌电信号,最后利用肌肉激活特性调制FES输出。实验表明开环肌电调制模式下的脚尖俯仰角仅有17°左右,而在闭环调制模式下,脚尖俯仰角最大角度达到了21°左右。本文设计的FES控制系统可以帮助足下垂患者进行康复训练。     

蓝牙无线表面肌电采集系统设计及应用

介绍一种基于MSP430的表面肌电蓝牙采集系统及其在膝关节病人肌肉功能检测中的应用。该系统包括信号采集与放大、A/D转换、蓝牙传输,实现了肌电信号采集与上位机的蓝牙通信,并成功用于采集膝关节病人下蹲动作的肌电信号。     

基于双谱分析的表面肌电信号模式识别

本文提出一种新的用于表面肌电信号分类的方法。这种方法将双谱分析技术应用于表面肌电信号分类来对六种简单的动作进行分类,包括内翻,外翻,握拳,展拳,上切和下切六种动作模式的识别。在以往的表面肌电信号分类中,人们都假设信号满足高斯分布和线性,并且为平稳信号。但是实际的表面肌电信号往往不能满足上面的假设,根据前人对表面肌电信号的研究我们知道,当肌肉收缩低于最大自发收缩的25％时,表面肌电信号所表现的非高斯性是显著的。因此为了获得更多的表面肌电信号的信息和获取更好的表面肌电信号分类的识别率,我们利用双谱分析和主元分析相结合方法对肌电信号进行了分类研究。     

三自由度比例控制肌电假手的设计

针对电动假手的仿生控制问题,给出了一种三自由度实时比例控制肌电假手的设计方案。通过采集残臂上的4路表面肌电信号（SEMG）,采用能量时域分析法对信号进行了特征提取,并采用二叉决策树模式分类方法识别得到了手部3个自由度7个动作模式。由单片机构成的信号处理和控制电路,实现了三自由度电动假手的实时仿生控制;并根据表面肌电信号的强弱,采用多路SEMG能量加权法来求取多自由度假手的比例控制系数,实现了对电动假手的比例控制。试验结果表明,基于表面肌电信号的三自由度肌电假手响应速度快。动作准确率达到96％以上。     

基于生物力学特性仿人机器人的研究

本文在介绍国内外仿人形机器人研究现状的基础上,基于人体舒适与疲劳的生物力学机理,分析了生物力学指标(肌肉硬度,肌电信号,血流特性)与人体各关节运动转角和人机接触面体压分布的关系,提出了人体潜意识动作姿态舒适度指标和人机接触界面舒适度指标,并依据这些指标对机器人的人体潜意识动作姿态的生成方法,路径规划方法以及本体构建中的运动机构设计与控制系统设计展开了讨论,对于国内基于生物力学特性的仿人机器人的研究具有推动作用.     

基于预抓取模式识别的假手肌电控制方法

为解决多自由度假手肌电控制难题,建立一种人手预抓取模式的在线识别方法,并将其应用至HIT-DLR假手的抓取控制。基于Teager-Kaiser能量算子增幅肌电信号在肌肉动作发起时的变化,引入后处理解决噪声影响,提出一种预抓取发起的在线检测方法。针对人手4种预抓取模式,讨论不同肌电信号分段方法,不同时域特征、频域特征和时频域特征以及多种分类方法所能获得的识别成功率。最终建立了基于波形长度及支持矢量机的最优识别方法,成功率可达95%,延迟小于300 ms。肌电控制试验表明,假手可以准确快速地抓取各种不同形状的物体。     

A novel estimating method for steering efficiency of the driver with electromyography signals

The existing research of steering efficiency mainly focuses on the mechanism efficiency of steering system, aiming at designing and optimizing the mechanism of steering system. In the development of assist steering system especially the evaluation of its comfort, the steering efficiency of driver physiological output usually are not considered, because this physiological output is difficult to measure or to estimate, and the objective evaluation of steering comfort therefore cannot be conducted with movement efficiency perspective. In order to take a further step to the objective evaluation of steering comfort, an estimating method for the steering efficiency of the driver was developed based on the research of the relationship between the steering force and muscle activity. First, the steering forces in the steering wheel plane and the electromyography （EMG） signals of the primary muscles were measured. These primary muscles are the muscles in shoulder and upper ann which mainly produced the steering torque, and their functions in steering maneuver were identified previously. Next, based on the multiple regressions of the steering force and EMG signals, both the effective steering force and the total force capacity of driver in steering maneuver were calculated. Finally, the steering efficiency of driver was estimated by means of the estimated effective force and the total force capacity, which represented the information of driver physiological output of the primary muscles. This research develops a novel estimating method for driver steering efficiency of driver physiological output, including the estimation of both steering force and the force capacity of primary muscles with EMG signals, and will benefit to evaluate the steering comfort with an objective perspective.     

Estimation of muscle and joint forces in the human lower extremity during rising motion from a seated position

Biomechanical models are often employed to predict in vivo muscle or joint forces in the human body because measuring these forces is difficult. Even though the rising motion from a seated position frequently occurs in daily life and the force acting on the knee joints during the motion is important for aged or infirmed people, limited studies related to the motion have been conducted. This study aims to propose a numerical procedure to estimate the muscle and joint forces in the human lower extremity during rising motion from a seated position. The human lower extremity is idealized as a multibody system in which the Hill-type muscle force model is employed. The multibody system consists of four bodies (shank, thigh, pelvis, and upper body), three revolute joints, and ten forces. The motion of the multibody system is assumed constrained to the sagittal plane, and the muscles in the human lower extremity are idealized by nine action/reaction forces. The nine forces are determined by minimizing the metabolic energy, which is consumed during the rising motion. Metabolic energy consists of the energy consumed by heat generation of muscles and the mechanical work done by muscles. For the accuracy validation of the proposed estimation method, numerical results obtained with the proposed method are compared with existing experimental results.     

基于改进小脑模型的sEMG下肢关节力矩预测

关节力矩预测在康复医学、临床医学和运动训练等领域有着重要作用,对力矩连续、实时地预测可以使人机交互设备更好地反馈、复刻人体运动意图。为了给患者提供一个安全、主动、舒适的康复训练环境,提升人机交互设备的柔顺性,提出了一种改进型递归小脑模型神经网络模型关节力矩预测方法。该方法采用肌肉协同分析对采集的相关肌肉的表面肌电信号(sEMG)进行降维,将降维后的sEMG特征向量与关节角速度、关节角度作为输入信号,并在小脑模型神经网络中加入递归单元和模糊逻辑规则,以小波函数作为隶属度函数,对非疲劳、过渡疲劳及疲劳这3种状态下的踝关节背屈跖屈运动的动态力矩进行连续预测。力矩预测值与实际值之间的平均皮尔逊相关系数和平均标准均方根误差分别为0.933 5和0.159 8,实验结果验证了该方法对下肢关节力矩连续预测的准确性和有效性。     

基于肌肉协同激活模型的上肢关节运动连续估计

关节运动连续估计为基于表面肌电信号的人机交互提供了一种更为自然灵活的方式。提出了一种基于肌肉协同理论和支持向量回归的激活模型进行上肢关节角度的估计。首先利用非负矩阵分解算法对肌电信号进行解耦,提取独立动作的协同元;然后根据非负最小二乘算法计算相应协同元激活系数;最后通过支持向量回归构建了映射激活系数到关节角度的激活模型,利用建立的激活模型从采集的表面肌电信号得到关节运动的连续估计。对2个关节独立和组合运动的估计实验表明,该模型能获得较高的估计精度。     

基于气动肌腱驱动的拮抗式仿生关节设计与控制

关节是仿生机器人机械系统的基本组成元素。在面向环境交互的仿生机器人系统中,关节的质量、体积以及功/重比直接影响系统的工作性能。与传统的电气、液压驱动方式相比,采用气动肌腱的驱动方式具有质量小、体积小和高功/重比等优势,具有广阔的应用前景。从功能仿生角度出发,在分析人体肘关节骨骼结构特点和肌肉发力方式的基础上,设计一种基于气动肌腱驱动的拮抗式仿生关节。建立单根肌腱的数学模型,通过搭建气动肌腱工作特性试验平台对肌腱进行性能测试,采用最小二乘法对模型参数进行辨识。针对拮抗式仿生关节的构型进行运动学和动力学分析,提出基于肌腱模型的偏置输入气压控制方法,设计基于关节估计阻尼的扰动观测器对名义模型进行补偿,通过数值仿真对控制方法进行有效性验证。建立仿生关节运动控制试验系统,通过单关节轨迹跟踪试验验证了肌腱理论模型的正确性及控制策略的有效性。     

一种基于单通道 sEMG 分解与 LSTM 神经网络 相结合的手势识别方法

在基于表面肌电(sEMG)信号的动作识别中,使用单通道传感器能够简化系统、减少识别延时,但也存在识别精度偏低的问题。为了提高识别精度,本文提出将单通道sEMG信号分解策略与长短期记忆(LSTM)循环神经网络识别相结合的方法。在该方法中,先将单通道sEMG信号分解成多通道运动单元动作电位序列(MUAPTs),然后提取MUAPTs的特征,最后将这些特征对LSTM分类模型进行训练。为了验证该方法的有效性,本文以手势动作识别为对象,对6名受试者分别建立了4种分类模型,包括基于未分解信号的支持向量机(SVM)、基于分解信号的SVM、基于未分解信号的LSTM、以及本文提出的基于分解信号的LSTM,并定义识别精度量化指标对这四种模型的分类结果进行评估。对于旋前方肌sEMG信号,在使用本文所提方法进行手势识别时,平均估计精度均能达到90%以上,比未分解的LSTM高18.7%,比分解信号的SVM高4.17%,比未分解信号的SVM高11.53%。实验结果验证了本文所提方法的有效性。     

改进 S 变换谱估计的动态手势肌电特征处理

手语是各种手势动态变化的一种复杂运动模式,手势特征处理效果直接关系到手语识别的准确性。本文提出一种基于改进S变换谱估计的动态手势肌电特征处理新方法。对采集的表面肌电信号进行S变换,引入优化因子调节时频分辨率并生成改进S变换谱;定义谱的时间和频率分量为二维随机变量,以改进S变换谱元素为二维随机变量样本,通过高斯核密度估计得到二维核密度函数。仿真和实验均表明,改进S变换谱估计方法有效抑制了白噪声,并使动态手势的肌电暂态突变特征得到加强。与经验模态分解、自排序熵、奇异值排序熵等方法对比,基于该方法的动态手势识别率分别提高了10.0%、6.67%和11.67%,特征处理方法的效果明显。     

基于sEMG与足底压力信号融合的跌倒检测研究

跌倒已经成为一种普遍危害老年人身心健康的事故,需要得到及时救治。设计了一种基于表面肌电(s EMG)和足底压力信号融合的跌倒检测系统。提取s EMG的近似熵及基本尺度熵特征,并根据足底压力的变化规律,提取动作信号段的压力特征,通过D-S证据推理将肌电信号与足底压力信号的SVM决策融合获得综合判别结果。实验结果表明,该方法对跌倒与ADL的平均识别率达到了91.7%,优于单一信源识别结果。     

Finger Flexion Motion Inference from sEMG Signals

This paper provides a method to infer finger flexing motions using a 4-channel surface Electromyogram(sEMG).Surface EMGs are harmless to the human body and easily done.However,they do not reflect the activity of specific nerves or muscles,unlike invasive EMGs.On the other hand,the non-invasive type is difficult to use for discriminating various motions while using only a small number of electrodes.Surface EMG data in this study were obtained from four electrodes placed around the forearm.The motions were the flexion of each 5 single fingers(thumb,index finger,middle finger,ring finger,and little fingers).One subject was trained with these motions and another left was untrained.The maximum likelihood estimation method was used to infer the finger motion.Experimental results have showed that this method could be useful for recognizing finger motions.The average accuracy was as high as 95%.     

Optimization of a quantitative method for muscle histology assessment

Optical microscopy offers the simplest way to obtain magnified images of biological tissues. The assessment of the muscle destructuration level can be performed by a method called Meat Destruction Indicator (MDI), which combines optical microscopy and image analysis. MDI can be used for evaluation of food quality and for considering mechanically separated meat (meat raw material with an MDI value above 58.1% contained muscle fibres sufficiently destructured). This paper is particularly focused on the metrological optimization of a quantitative image analysis method around the example of MDI measurement by microscopy, especially on the digital acquisition calibration focusing and analysis work‐flow. Ten different samples (45 sections) were examined with variable settings of microscope and camera to define the optimal configuration. The tests were performed with different observers to define rules and criteria for results validation. Based on the obtained results, we suggest choosing objective rules to set the light and colour of the camera and the microscope focus. To control the results of the automatic segmentation emerged also as a key step, and objective rules for observers to select or discard wrong segmented images should be defined. The adjusted MDI measurement by microscope can be used as a reliable method with good repeatability, thanks to this metrological assessment, which could and should be applied to all image analysis applications whatever the application.