基于表面肌电的意图识别方法在非理想条件下的研究进展

李自由; 赵新刚; 张弼; 丁其川; 张道辉; 韩建达

doi:10.16383/j.aas.c200263

基于表面肌电的意图识别方法在非理想条件下的研究进展

doi: 10.16383/j.aas.c200263

李自由^{1, 2, 3},
赵新刚^{1, 2},
张弼^{1, 2},
丁其川⁴,
张道辉^{1, 2},
韩建达^{1, 2, 5}

1.
中国科学院沈阳自动化研究所机器人学国家重点实验室沈阳 110016
2.
中国科学院机器人与智能制造创新研究院沈阳 110016
3.
中国科学院大学北京 100049
4.
东北大学机器人科学与工程学院沈阳 110004
5.
南开大学人工智能学院天津 300350

基金项目: 国家自然科学基金(61773369, U1813214)、中国博士后科学基金项目(2019M661157)资助

详细信息

作者简介:
李自由：中国科学院沈阳自动化研究所博士研究生, 主要研究方向为生物电信号处理, 模式识别与机器学习

赵新刚：中国科学院沈阳自动化研究所研究员. 主要研究方向为机器人控制, 智能系统与康复机器人

张弼：中国科学院沈阳自动化研究所副研究员, 主要研究方向为先进控制理论及应用, 智能机器人交互

丁其川：东北大学机器人科学与工程学院副教授. 主要研究方向为人体运动意图识别和智能机器人

张道辉：中国科学院沈阳自动化研究所副研究员, 主要研究方向为机器人控制技术

韩建达：南开大学人工智能学院教授.主要研究方向为智能系统, 移动机器人自主控制

计量
- 文章访问数: 25
- HTML全文浏览量: 11
- 被引次数: 0
出版历程
- 收稿日期: 2020-04-29
- 录用日期: 2020-12-14
- 网络出版日期: 2021-01-19

Review of sEMG-based Motion Intent Recognition Methods in Non-ideal Conditions

LI Zi-You^{1, 2, 3},
ZHAO Xin-Gang^{1, 2},
ZHANG Bi^{1, 2},
DING Qi-Chuan⁴,
ZHANG Dao-Hui^{1, 2},
Han Jian-Da^{1, 2, 5}

1.
Skate Key Laboratory of Robotics, Shenyang Institute of Automation, Chinese Academy of Sciences, Shenyang 110016
2.
Institute of Robotics and Intelligent Manufacturing Innovation, Chinese Academy of Sciences, Shenyang 110016
3.
University of Chinese Academy of Sciences, Beijing 100049
4.
School of Robotics and Engineering, Northeastern University, Shenyang 110004
5.
School of Artificial Intelligence, Nankai University, Tianjin 300350

Funds: Supported by National Science Foundation of China (61773369, U1813214), the China Postdoctoral Science Foundation (No. 2019M661157)

摘要

摘要: 在基于表面肌电信号(surface electromyogram, sEMG)的意图识别研究领域, 目前大多数的研究主要集中在提高肌电识别的准确性方面. 然而, 在实际应用中, 基于sEMG识别的交互系统往往受到诸多非理想因素干扰, 肌电识别的准确性被大大降低. 本文主要关注在非理想条件下肌电识别的鲁棒性研究, 首先详细归纳了肌电识别方法受到的非理想干扰因素(如电极偏移、个体性差异、肌肉疲劳、肢体姿态或其他综合性干扰), 总结了当前研究的抗干扰方法; 随后讨论了非理想干扰因素研究现状中的主要问题; 最后在构建肌电数据集、探索深度学习和迁移学习, 以及肌电分解研究等方面, 对未来的关键技术进行了展望.
- 肌电信号 /
- 研究进展 /
- 非理想条件 /
- 模式识别
Abstract: In sEMG-based recognition, most studies are currently focusing on improving recognition accuracies. While in real applications, sEMG-based recognition systems are limited by many disturbances in non-ideal conditions, and recognition accuracies are worsen greatly. This paper is focusing on the robustness of sEMG-based recognition. Many disturbances in non-ideal conditions are detailed and summarized, including electrode shifts, individual differences, muscle fatigue, limb postures and others. Also, many novel methods that are proposed to remove or reduce the impact of these disturbances are summarized. Furthermore, main problems in these current studies are discussed. Finally, prospections for the future development are proposed, including building sEMG-based datasets, exploiting deep learningbased and transfer learning-based recognition, and sEMG decomposition.
- electromyogram-based recognition /
- researches and prospects /
- non-ideal conditions /
- pattern recognition

HTML全文

图 1 基于监督学习的sEMG识别模型训练与测试框架

Fig. 1 The training and testing framework of sEMG recognition model based on supervised learning

下载: 全尺寸图片幻灯片

图 2 (a)稀疏电极不同位置^[13], (b)环形电极旋转偏移^[15], 和(c)HD-sEMG电极中的纵向偏移和横向偏移^[14]

Fig. 2 (a) Different positions of sparse electrodes^[13], (b) Ring electrode rotation offset^[15], (c) Longitudinal and transverse offsets in HD-sEMG electrodes^[14]

下载: 全尺寸图片幻灯片

图 3 基于用户依赖因子(User-dependent factor)和动作依赖因子(Motion-dependent factor)构建用户无关特征(User-independent feature)的双线性模型^[45]

Fig. 3 User-independent bilinear model based on user-dependent factor and motion-dependent factor^[45]

下载: 全尺寸图片幻灯片

图 4 疲劳状态下肌电信号中值频率与肌肉输出力变化^[52]

Fig. 4 Changes of median frequency of EMG signal and muscle output force under fatigue condition^[52]

下载: 全尺寸图片幻灯片

图 5 (a)不同的静态姿态, (b)模拟日常活动中的肢体姿态和(c)肢体姿态组合动态运动^[70]

Fig. 5 (a) Different static posture, (b) simulating the dynamic movement of body posture and (c) combination of body posture in daily activities^[70]

下载: 全尺寸图片幻灯片

图 6 非理想肌电的关键技术展望

Fig. 6 The key technology prospects of non-ideal EMG

下载: 全尺寸图片幻灯片

图 7 基于TCN网络结构的肌电信号时序、层级特征提取框架^[96]

Fig. 7 Sequential and hierarchical feature extraction framework of EMG signal based on TCN network^[96]

下载: 全尺寸图片幻灯片

图 8 基于迁移学习的理想场景与非理想场景之间肌电识别模型的更新与适应

Fig. 8 Update and adaptation of sEMG-based recognition model between ideal and non-ideal scenarios based on transfer learning

下载: 全尺寸图片幻灯片

图 9 sEMG分解与MUAP^[107]

Fig. 9 Surface EMG signal decomposition and MUAP^[107]

下载: 全尺寸图片幻灯片

表 1 非理想因素及解决方案

Table 1 Non-ideal factors and solutions

干扰\方法	数据扩增	鲁棒特征	模型更新
电极偏移	1.多位置数据扩增^[13,16]; 2.多通道数据扩增^[12]; 3.惯导等数据融合^[17,18];	1.AR、TDAR特征^[16,19]; 2.倒频谱特征^[20]; 3.Variogram特征^[10]; 4.结构相似性特征^[21]; 5.共空间模式特征^[22]; 6.灰度共生矩阵特征^[14];	1.模型修正与协方差偏移适应^[24,25] 2.期望最大化迁移学习^[15,26,27] 3.自适应增量式混合分类器^[28,29]; 4.偏移估计与模型更新^[30]; 5.骨骼位置估计与校正^[31];
个体性差异	1.多人数据扩增^[36]; 2.惯导等数据融合^[37]; 3.个体形态参数归一化^[38];	1.肌电分解特征^[39]; 2.多分辨率肌肉协同特征^[40]; 3.共同空间映射^[41];	1.SVM权重更新策略^[42,43,90] 2.典型相关分析低维共空间映射^[44] 3.用户、动作依赖的双线性模型^[45] 4.基于最大收缩力的模型泛化^[46] 5.基于卷积神经网络的模型迁移^[47]
肢体姿态	1.多姿态下的数据扩增^[75–77]; 2.“动态训练”数据采集^[70,78]; 3.惯导等数据融合^{[17,75,76,79]};	1. 谱矩等频域特征^[80–82]; 2. 稀疏表达特征^[83,84];	—
肌肉疲劳	1.频域特征归一化(MDF、MNF^[52,91]、STFT、WT^[92,93]、iMDF、iMNF^[56]、一维频谱-标准差^[59]等); 2.疲劳状态的分类识别^[59,62–64];

下载: 导出CSV

表 2 sEMG数据集

Table 2 Surface EMG signal datasets

	数据集	传感器	参与人数	动作类别
NinaPro^[4]	DB1	1.Otto Bock MyoBock 13E200电极, (10通道, 肌电RMS特征) 2.数据手套Cyberglove(22通道)	27健康人	52类手部动作重复10次
	DB2/3	1. Delsys Trigno Wireless 肌电系统, (12通道双差分EMG, 36通道ACC) 2. 数据手套Cyberglove (22通道) 3. 手指力传感器(6通道)4. 腕部倾角传感器(2通道)	40健康人	49个手部动作重复6次
	DB4	1. Cometa单差分无线肌电电极(12 通道单差分)	10名健康人	52类手部动作重复6次
	DB5	1. Thalmic Myo肌电臂环(2套, 共计 2×8通道单差分) 2. 数据手套Cyberglove(22通道)	10名健康人	52类手部动作重复6次
	DB6	1. Delsys Trigno Wireless肌电系统 (14通道, 42通道ACC) 2. Tobii Pro Glasses II(追踪眼动和视野)	10名健康人	7个抓握动作重复12次重复5天
	DB7	1. Delsys Trigno Wireless 肌电系统 (12通道EMG, 9轴IMU) 2. 数据手套Cyberglove(22通道)	20名健康人2名截肢者	40个手部动作重复6次
	DB8	1. Delsys Trigno Wireless 肌电系统 (16通道EMG, 9轴IMU, 采样至2 kHz) 2. 数据手套Cyberglove (22通道)	10名健康人2名截肢者	9个手部动作
CSL-HDEMG^[31]		1. HD-EMG(8×24=192通道)	5名健康人	27个手部/手指动作
CapgMyo^[95]	DB-a	1. HD-EMG(8x16=128通道)	18名健康人	8个手指动作 8个手指动作(同上) 不同时间段重复两次 12个手指动作
	DB-b		10名健康人
	DB-c		10名健康人
UCI等其他	Myo^[108]	1. Thalmic Myo肌电臂环(8通道)	36名健康人	8个抓握动作
UCI等其他	Christos-Delsys^[109]	1. Delsys Trigno Wireless (2通道)	6名健康人	6个抓握动作

下载: 导出CSV

[1]	丁其川, 熊安斌, 赵新刚, 韩建达. 基于表面肌电的运动意图识别方法研究及应用综述. 自动化学报, 2016, 42(1): 13−25 Ding Qi-Chuan, Xiong An-Bin, Zhao Xin-Gang, Han JianDa. A review on researches and applications of sEMG-based motion intent recognition methods. Acta Automatica Sinica, 2016, 42(1): 13−25
[2]	J. He and N. Jiang. Biometric from surface electromyogram (semg): Feasibility of user verification and identification based on gesture recognition. Frontiers in Bioengineering and Biotechnology, 2020, 8: 58 doi: 10.3389/fbioe.2020.00058
[3]	赵新刚, 谈晓伟, 张弼. 柔性下肢外骨骼机器人研究进展及关键技术分析. 机器人, 2020, 42(3): 365−384 ZHAO Xingang, TAN Xiaowei, ZHANG Bi. Development of Soft Lower Extremity Exoskeleton and Its Key Technologies: A Survey. ROBOT, 2020, 42(3): 365−384
[4]	M. Atzori, A. Gijsberts, C. Castellini, B. Caputo, A.-G.M. Hager, S. Elsig, G. Giatsidis, F. Bassetto, and H. Muller. Electromyography data for non-invasive ¨ naturally-controlled robotic hand prostheses. Scientific Data, 2014, 1: 140053 doi: 10.1038/sdata.2014.53
[5]	M. Cognolato, A. Gijsberts, V. Gregori, G. Saetta, K. Giacomino, A.-G. M. Hager, A. Gigli, D. Faccio, C. Tiengo, F. Bassetto, et al. Gaze, visual, myoelectric, and inertial data of grasps for intelligent prosthetics. Scientific Data, 2020, 7(1): 1−15 doi: 10.1038/s41597-019-0340-y
[6]	丁其川, 赵新刚, 韩建达. 基于肌电信号的上肢多关节连续运动估计. 机器人, 2014, 36(4): 469−476 Ding Qi-Chuan, Zhao Xin-Gang, Han Jian-Da. EMG-based estimation for multi-joint continuous movement of human upper limb. Robot, 2014, 36(4): 469−476
[7]	A. Belyea, K. Englehart, and E. Scheme. Fmg versus emg: A comparison of usability for real-time pattern recognition based control. IEEE Transactions on Biomedical Engineering, 2019, 66(11): 3098−3104 doi: 10.1109/TBME.2019.2900415
[8]	L. Bi, C. Guan, et al. A review on emg-based motor intention prediction of continuous human upper limb motion for human-robot collaboration. Biomedical Signal Processing and Control, 2019, 51: 113−127 doi: 10.1016/j.bspc.2019.02.011
[9]	B. Hudgins, P. Parker, and R. N. Scott. A new strategy for multifunction myoelectric control. IEEE Trans Biomed Eng, 1993, 40(1): 82−94 doi: 10.1109/10.204774
[10]	A. Stango, F. Negro, and D. Farina. Spatial correlation of high density emg signals provides features robust to electrode number and shift in pattern recognition for myocontrol. IEEE Trans Neural Syst Rehabil Eng, 2015, 23(2): 189−98 doi: 10.1109/TNSRE.2014.2366752
[11]	A. J. Young, L. J. Hargrove, and T. A. Kuiken. The effects of electrode size and orientation on the sensitivity of myoelectric pattern recognition systems to electrode shift. IEEE Trans Biomed Eng, 2011, 58(9): 2537−44 doi: 10.1109/TBME.2011.2159216
[12]	J. He, X. Sheng, X. Zhu, and N. Jiang. Electrode density affects the robustness of myoelectric pattern recognition system with and without electrode shift. IEEE J Biomed Health Inform, 2019, 23(1): 156−163 doi: 10.1109/JBHI.2018.2805760
[13]	L. Hargrove, K. Englehart, and B. Hudgins. The effect of electrode displacements on pattern recognition based myoelectric control. Conf Proc IEEE Eng Med Biol Soc, 2006, 1: 2203−6
[14]	J. He and X. Zhu. Combining improved gray-level co-occurrence matrix with high density grid for myoelectric control robustness to electrode shift. IEEE Trans Neural Syst Rehabil Eng, 2017, 25(9): 1539−1548 doi: 10.1109/TNSRE.2016.2644264
[15]	C. Prahm, A. Schulz, B. Paaben, J. Schoisswohl, E. Kaniusas, G. Dorffner, B. Hammer, and O. Aszmann. Counteracting electrode shifts in upper-limb prosthesis control via transfer learning. IEEE Trans Neural Syst Rehabil Eng, 2019, 27(5): 956−962 doi: 10.1109/TNSRE.2019.2907200
[16]	L. Hargrove, K. Englehart, and B. Hudgins. A training strategy to reduce classification degradation due to electrode displacements in pattern recognition based myoelectric control. Biomedical Signal Processing and Control, 2008, 3(2): 175−180 doi: 10.1016/j.bspc.2007.11.005
[17]	A. Fougner, E. Scheme, A. D. Chan, K. Englehart, and O. Stavdahl, “A multi-modal approach for hand motion classification using surface emg and accelerometers, ” Conf Proc IEEE Eng Med Biol Soc, pp. 4247–50, 2011.
[18]	N. M. Lopez, F. di Sciascio, C. M. Soria, and M. E. Valentinuzzi. Robust emg sensing system based on data fusion for myoelectric control of a robotic arm. Biomed Eng Online, 2009, 8: 5 doi: 10.1186/1475-925X-8-5
[19]	A. J. Young, L. J. Hargrove, and T. A. Kuiken. Improving myoelectric pattern recognition robustness to electrode shift by changing interelectrode distance and electrode configuration. IEEE Trans Biomed Eng, 2012, 59(3): 645−52 doi: 10.1109/TBME.2011.2177662
[20]	D. Tkach, H. Huang, and T. A. Kuiken. Study of stability of time-domain features for electromyographic pattern recognition. J Neuroeng Rehabil, 2010, 7: 21 doi: 10.1186/1743-0003-7-21
[21]	A. Boschmann and M. Platzner, “Towards robust hd emg pattern recognition: reducing electrode displacement effect using structural similarity, ” Conf Proc IEEE Eng Med Biol Soc, pp. 4547–50, 2014.
[22]	L. Pan, D. Zhang, N. Jiang, X. Sheng, and X. Zhu. Improving robustness against electrode shift of high density emg for myoelectric control through common spatial patterns. J Neuroeng Rehabil, 2015, 12: 110 doi: 10.1186/s12984-015-0102-9
[23]	G. Huang, Z. Xian, F. Tang, L. Li, L. Zhang, and Z. Zhang. Low-density surface electromyographic patterns under electrode shift: Characterization and nmf-based classification. Biomedical Signal Processing and Control, 2020, 59: 101890 doi: 10.1016/j.bspc.2020.101890
[24]	S. Amsuss, P. M. Goebel, N. Jiang, B. Graimann, L. Paredes, and D. Farina. Self-correcting pattern recognition system of surface emg signals for upper limb prosthesis control. IEEE Trans Biomed Eng, 2014, 61(4): 1167−76 doi: 10.1109/TBME.2013.2296274
[25]	M. M. Vidovic, H. J. Hwang, S. Amsuss, J. M. Hahne, D. Farina, and K. R. Muller. Improving the robustness of myoelectric pattern recognition for upper limb prostheses by covariate shift adaptation. IEEE Trans Neural Syst Rehabil Eng, 2016, 24(9): 961−970 doi: 10.1109/TNSRE.2015.2492619
[26]	C. Prahm, B. Paassen, A. Schulz, B. Hammer, and O. Aszmann. Transfer learning for rapid re-calibration of a myoelectric prosthesis after electrode shift. Converging Clinical and Engineering Research on Neurorehabilitation Ii, Vols 1 and 2, 2017, 15: 153−157
[27]	B. Paaßen, A. Schulz, J. Hahne, and B. Hammer. Expectation maximization transfer learning and its application for bionic hand prostheses. Neurocomputing, 2018, 298: 122−133 doi: 10.1016/j.neucom.2017.11.072
[28]	丁其川, 赵新刚, 李自由, 韩建达. 基于自更新混合分类器模型的肌电运动识别方法. 自动化学报, 2018, 45(8): 1464−1474 Ding Qi-Chuan, Zhao Xin-Gang, Li Zi-You, Han Jian-Da. The EMG-motion recognition method based on self-update hybrid classification model. Acta Automatica Sinica, 2018, 45(8): 1464−1474
[29]	Q. Ding, X. Zhao, J. Han, C. Bu, and C. Wu. Adaptive hybrid classifier for myoelectric pattern recognition against the interferences of outlier motion, muscle fatigue, and electrode doffing. IEEE Trans Neural Syst Rehabil Eng, 2019, 27(5): 1071−1080 doi: 10.1109/TNSRE.2019.2911316
[30]	李自由, 王丰焱, 赵新刚, 丁其川, 张道辉, 韩建达. 基于Myo旋转偏移估计与自适应校正的手势识别方法. 自动化学报, 2020, 46(9): 1896−1907 Li Zi-You, Wang Feng-Yan, Zhao Xin-Gang, Ding Qi-Chuan, Zhang Dao-Hui, Han Jian-Da. The method for gestures recognition based on Myo rotation shifts estimation and adaptive correction. Acta Automatica Sinica, 2020, 46(9): 1896−1907
[31]	C. Amma, T. Krings, J. Boer, and T. Schultz, “Advancing ¨ muscle-computer interfaces with high-density electromyography, ” pp. 929–938, 2015.
[32]	A. Ameri, M. A. Akhaee, E. Scheme, and K. Englehart. A deep transfer learning approach to reducing the effect of electrode shift in emg pattern recognition-based control. IEEE Transactions on Neural Systems and Rehabilitation Engineering, 2019, 28(2): 370−379
[33]	G. Li, O. W. Samuel, C. Lin, M. G. Asogbon, P. Fang, and P. O. Idowu, “Realizing efficient emg-based prosthetic control strategy, ” in Neural Interface: Frontiers and Applications. Springer, 2019, pp. 149–166.
[34]	E. Campbell, A. Phinyomark, A. H. Al-Timemy, R. N. Khushaba, G. Petri, and E. Scheme, “Differences in emg feature space between able-bodied and amputee subjects for myoelectric control, ” in 2019 9th International IEEE/EMBS Conference on Neural Engineering (NER). IEEE, 2019, pp. 33–36.
[35]	K.-T. Kim, C. Guan, and S.-W. Lee. A subject-transfer framework based on single-trial emg analysis using convolutional neural networks. IEEE Transactions on Neural Systems and Rehabilitation Engineering, 2019, 28(1): 94−103
[36]	G. Weichao, S. Xinjun, L. Jianwei, H. Lei, Z. Dingguo, and Z. Xiangyang, “Towards zero training for myoelectric control based on a wearable wireless semg armband, ” 2015 IEEE International Conference on Advanced Intelligent Mechatronics (AIM), pp. 196–201, 2015.
[37]	F. Stival, S. Michieletto, A. De Agnoi, and E. Pagello, “Toward a better robotic hand prosthesis control: Using emg and imu features for a subject independent multi joint regression model, ” pp. 185–192, 2018.
[38]	A. Phinyomark, F. Quaine, S. Charbonnier, C. Serviere, F. Tarpin-Bernard, and Y. Laurillau. A feasibility study on the use of anthropometric variables to make muscle–computer interface more practical. Engineering Applications of Artificial Intelligence, 2013, 26(7): 1681−1688 doi: 10.1016/j.engappai.2013.01.004
[39]	X. Anbin, Z. Xingang, H. Jianda, G. Liu, and D. Qichuan, “An user-independent gesture recognition method based on semg decomposition, ” 2015 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pp. 4185–4190, 2015.
[40]	M. R. Ison and P. Artemiadis, “Beyond user-specificity for emg decoding using multiresolution muscle synergy analysis, ” Proceedings of the ASME 2013 Dynamic Systems and Control Conference (DSCC2013), 2013.
[41]	S. P. Rita Chattopadhyay, Narayanan C Krishnan, “Topology preserving domain adaptation for addressing subject based variability in semg signal, ” 2011 AAAI Spring symposium series, 2011.
[42]	T. Tommasi, F. Orabona, C. Castellini, and B. Caputo. Improving control of dexterous hand prostheses using adaptive learning. IEEE Transactions on Robotics, 2013, 29(1): 207−219 doi: 10.1109/TRO.2012.2226386
[43]	N. Patricia, T. Tommasit, and B. Caputo, “Multi-source adaptive learning for fast control of prosthetics hand, ” 2014 22nd International Conference on Pattern Recognition, pp. 2769–2774, 2014.
[44]	R. N. Khushaba. Correlation analysis of electromyogram signals for multiuser myoelectric interfaces. IEEE Trans Neural Syst Rehabil Eng, 2014, 22(4): 745−55 doi: 10.1109/TNSRE.2014.2304470
[45]	T. Matsubara and J. Morimoto. Bilinear modeling of emg signals to extract user-independent features for multiuser myoelectric interface. IEEE Trans Biomed Eng, 2013, 60(8): 2205−13 doi: 10.1109/TBME.2013.2250502
[46]	L. Pan, D. L. Crouch, and H. Huang. Myoelectric control based on a generic musculoskeletal model: Toward a multi-user neural-machine interface. IEEE Transactions on Neural Systems and Rehabilitation Engineering, 2018, 26(7): 1435−1442 doi: 10.1109/TNSRE.2018.2838448
[47]	U. Cote-Allard, C. L. Fall, A. Campeau-Lecours, C. Gosselin, F. Laviolette, and B. Gosselin, “Transfer learning for semg hand gestures recognition using convolutional neural networks, ” 2017 IEEE International Conference on Systems, Man, and Cybernetics (Smc), pp. 1663–1668, 2017.
[48]	S. Cobb and A. Forbes. Electromyographic studies of muscular fatigue in man. American Journal of Physiology-Legacy Content, 1923, 65(2): 234−251 doi: 10.1152/ajplegacy.1923.65.2.234
[49]	R. M. Enoka and J. Duchateau. Muscle fatigue: what, why and how it influences muscle function. J Physiol, 2008, 586(1): 11−23 doi: 10.1113/jphysiol.2007.139477
[50]	K. Kim, Z. Sheng, and N. Sharma. Assessing fes-induced muscle fatigue using ultrasound to determine the inverse neuromuscular model for optimal fes input. The Journal of the Acoustical Society of America, 2019, 145(3): 1858−1858
[51]	H. A. Yousif, A. Zakaria, N. A. Rahim, A. F. B. Salleh, M. Mahmood, K. A. Alfarhan, L. M. Kamarudin, S. M. Mamduh, A. M. Hasan, and M. K. Hussain, “Assessment of muscles fatigue based on surface emg signals using machine learning and statistical approaches: A review, ” in IOP Conference Series: Materials Science and Engineering, vol. 705, no. 1. IOP Publishing, 2019, p. 012010.
[52]	C. J. De Luca. Myoelectrical manifestations of localized muscular fatigue in humans. Crit Rev Biomed Eng, 1984, 11(4): 251−279
[53]	R. Merletti, M. Knaflitz, and C. J. De Luca. Myoelectric manifestations of fatigue in voluntary and electrically elicited contractions. J Appl Physiol (1985), 1990, 69(5): 1810−20 doi: 10.1152/jappl.1990.69.5.1810
[54]	M. Cifrek, V. Medved, S. Tonkovic, and S. Ostojic. Surface emg based muscle fatigue evaluation in biomechanics. Clin Biomech (Bristol, Avon), 2009, 24(4): 327−40 doi: 10.1016/j.clinbiomech.2009.01.010
[55]	S. E. Jero and S. Ramakrishnan, “Analysis of muscle fatigue conditions in surface emg signal with a novel hilbert marginal spectrum entropy method, ” in 2019 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC). IEEE, 2019, pp. 2675–2678.
[56]	P. Bonato, S. H. Roy, M. Knaflitz, and C. J. De Luca. Time-frequency parameters of the surface myoelectric signal for assessing muscle fatigue during cyclic dynamic contractions. IEEE Trans Biomed Eng, 2001, 48(7): 745−53 doi: 10.1109/10.930899
[57]	H. Cao, I. El Hajj Dib, J. Antoni, and C. Marque. Analysis of muscular fatigue during cyclic dynamic movement. Conf Proc IEEE Eng Med Biol Soc, 2007, 2007: 1880−3
[58]	S. Thongpanja, A. Phinyomark, P. Phukpattaranont, and C. Limsakul. Mean and median frequency of emg signal to determine muscle force based on time-dependent power spectrum. Electronics and Electrical Engineering, 2013, 19(3)
[59]	M. R. Al-Mulla and F. Sepulveda. Novel feature modelling the prediction and detection of semg muscle fatigue towards an automated wearable system. Sensors (Basel), 2010, 10(5): 4838−54 doi: 10.3390/s100504838
[60]	L. Xu, X. Chen, S. Cao, X. Zhang, and X. Chen. A fatigue involved modification framework for force estimation in fatiguing contraction. IEEE Trans Neural Syst Rehabil Eng, 2018, 26(11): 2153−2164 doi: 10.1109/TNSRE.2018.2872554
[61]	X. Sheng, X. Ding, W. Guo, L. Hua, M. Wang, and X. Zhu, “Toward an integrated multi-modal semg/mmg/nirs sensing system for human-machine interface robust to muscular fatigue, ” IEEE Sensors Journal, 2020.
[62]	M. R. Al-Mulla, F. Sepulveda, M. Colley, and A. Kattan. Classification of localized muscle fatigue with genetic programming on semg during isometric contraction. Conf Proc IEEE Eng Med Biol Soc, 2009, 2009: 2633−8
[63]	Q. Wu, J. F. Mao, C. F. Wei, S. Fu, R. Law, L. Ding, B. T. Yu, B. Jia, and C. H. Yang. Hybrid bf–pso and fuzzy support vector machine for diagnosis of fatigue status using emg signal features. Neurocomputing, 2016, 173: 483−500 doi: 10.1016/j.neucom.2015.06.002
[64]	Q. Wu, C. Xi, L. Ding, C. Wei, H. Ren, R. Law, H. Dong, and X. L. Li. Classification of emg signals by bfa-optimized gsvcm for diagnosis of fatigue status. IEEE Transactions on Automation Science and Engineering, 2017, 14(2): 915−930 doi: 10.1109/TASE.2016.2564419
[65]	G. Jang, J. Kim, S. Lee, and Y. Choi. Emg-based continuous control scheme with simple classifier for electric-powered wheelchair. IEEE Transactions on Industrial Electronics, 2016, 63(6): 3695−3705 doi: 10.1109/TIE.2016.2522385
[66]	P. K. Artemiadis and K. J. Kyriakopoulos. A switching regime model for the emg-based control of a robot arm. IEEE Trans Syst Man Cybern B Cybern, 2011, 41(1): 53−63 doi: 10.1109/TSMCB.2010.2045120
[67]	T. D. Lalitharatne, K. Teramoto, Y. Hayashi, T. Nanayakkara, and K. Kiguchi. Evaluation of fuzzy-neuro modifiers for compensation of the effects of muscle fatigue on emg-based control to be used in upper-limb power-assist exoskeletons. Journal of Advanced Mechanical Design, Systems, and Manufacturing, 2013, 7(4): 736−751 doi: 10.1299/jamdsm.7.736
[68]	A. D. Enlena Mainardi, Ernesto Urbano, “Design of a new emg sensor for upper limb prosthetic control and real time frequency analysis, ” Proceedings of the 2008 MyoElectric Controls/Powered Prosthetics Symposium, 2008.
[69]	L. Peternel, N. Tsagarakis, D. Caldwell, and A. Ajoudani, “Adaptation of robot physical behaviour to human fatigue in human-robot co-manipulation, ” 2016 IEEE-RAS 16th International Conference on Humanoid Robots (Humanoids), pp. 489–494, 2016.
[70]	E. Scheme, K. Biron, and K. Englehart. Improving myoelectric pattern recognition positional robustness using advanced training protocols. Conf Proc IEEE Eng Med Biol Soc, 2011, 2011: 4828−4831
[71]	J. Liu, D. Zhang, J. He, and X. Zhu, “Effect of dynamic change of arm position on myoelectric pattern recognition, ” 2012 IEEE International Conference on Robotics and Biomimetics (ROBIO), pp. 1470–1475, 2012.
[72]	N. Jiang, S. Muceli, B. Graimann, and D. Farina. Effect of arm position on the prediction of kinematics from emg in amputees. Med Biol Eng Comput, 2013, 51(1-2): 143−51 doi: 10.1007/s11517-012-0979-4
[73]	D. Yang, W. Yang, Q. Huang, and H. Liu. Classification of multiple finger motions during dynamic upper limb movements. IEEE J Biomed Health Inform, 2017, 21(1): 134−141 doi: 10.1109/JBHI.2015.2490718
[74]	C. R. Beringer, M. Mansouri, L. E. Fisher, J. L. Collinger, M. C. Munin, M. L. Boninger, and R. A. Gaunt. The effect of wrist posture on extrinsic finger muscle activity during single joint movements. Scientific Reports, 2020, 10(1): 1−11 doi: 10.1038/s41598-019-56847-4
[75]	E. Scheme, A. Fougner, O. Stavdahl, A. C. Chan, and K. Englehart. Examining the adverse effects of limb position on pattern recognition based myoelectric control. Conf Proc IEEE Eng Med Biol Soc, 2010, 2010: 6337−40
[76]	A. Fougner, E. Scheme, A. D. Chan, K. Englehart, and O. Stavdahl. Resolving the limb position effect in myoelectric pattern recognition. IEEE Trans Neural Syst Rehabil Eng, 2011, 19(6): 644−51 doi: 10.1109/TNSRE.2011.2163529
[77]	K. H. Park, H. I. Suk, and S. W. Lee. Position-independent decoding of movement intention for proportional myoelectric interfaces. IEEE Trans Neural Syst Rehabil Eng, 2016, 24(9): 928−939 doi: 10.1109/TNSRE.2015.2481461
[78]	A. Radmand, E. Scheme, and K. Englehart. On the suitability of integrating accelerometry data with electromyography signals for resolving the effect of changes in limb position during dynamic limb movement. Journal of Prosthetics and Orthotics, 2014, 26(4): 185−193 doi: 10.1097/JPO.0000000000000041
[79]	Y. Geng, P. Zhou, and G. Li. Toward attenuating the impact of arm positions on electromyography pattern-recognition based motion classification in transradial amputees. J Neuroeng Rehabil, 2012, 9: 74 doi: 10.1186/1743-0003-9-74
[80]	R. N. Khushaba, S. Lei, and S. Kodagoda, “Time-dependent spectral features for limb position invariant myoelectric pattern recognition, ” 2012 International Symposium on Communications and Information Technologies (ISCIT), pp. 1015–1020, 2012.
[81]	R. N. Khushaba, M. Takruri, J. V. Miro, and S. Kodagoda. Towards limb position invariant myoelectric pattern recognition using time-dependent spectral features. Neural Networks, 2014, 55: 42−58 doi: 10.1016/j.neunet.2014.03.010
[82]	R. N. Khushaba, A. Al-Timemy, S. Kodagoda, and K. Nazarpour. Combined influence of forearm orientation and muscular contraction on emg pattern recognition. Expert Systems with Applications, 2016, 61: 154−161 doi: 10.1016/j.eswa.2016.05.031
[83]	J. L. Betthauser, C. L. Hunt, L. E. Osborn, R. R. Kaliki, and N. V. Thakor. Limb-position robust classification of myoelectric signals for prosthesis control using sparse representations. Conf Proc IEEE Eng Med Biol Soc, 2016, 2016: 6373−6376
[84]	J. L. Betthauser, C. L. Hunt, L. E. Osborn, M. R. Masters, G. Levay, R. R. Kaliki, and N. V. Thakor. Limb position tolerant pattern recognition for myoelectric prosthesis control with adaptive sparse representations from extreme learning. IEEE Trans Biomed Eng, 2018, 65(4): 770−778 doi: 10.1109/TBME.2017.2719400
[85]	丁其川, 赵新刚, 韩建达. 基于肌电信号容错分类的手部动作识别. 机器人, 2015, 37(1): 9−16 DING Qichuan, ZHAO Xingang, HAN Jianda. Recognizing Hand Motions Based on Fault-tolerant Classification with EMG Signals. ROBOT, 2015, 37(1): 9−16
[86]	M. Akmal, S. Zubair, M. Jochumsen, E. N. Kamavuako, and I. K. Niazi. A tensor-based method for completion of missing electromyography data. Ieee Access, 2019, 7: 104 710−104 720 doi: 10.1109/ACCESS.2019.2931371
[87]	D. Bell. The influence of air temperature on the emg/force relationship of the quadriceps. European journal of applied physiology and occupational physiology, 1993, 67(3): 256−260 doi: 10.1007/BF00864225
[88]	M. Ladrova, R. Martinek, J. Nedoma, and M. Fajkus, “Methods of power line interference elimination in emg signal, ” in Journal of Biomimetics, Biomaterials and Biomedical Engineering, vol. 40. Trans Tech Publ, 2019, pp. 64–70
[89]	Y. Zheng and X. Hu. Interference removal from electromyography based on independent component analysis. IEEE Transactions on Neural Systems and Rehabilitation Engineering, 2019, 27(5): 887−894 doi: 10.1109/TNSRE.2019.2910387
[90]	F. Orabona, C. Castellini, B. Caputo, A. E. Fiorilla, and G. Sandini, “Model adaptation with least-squares svm for adaptive hand prosthetics, ” 2009 IEEE International Conference on Robotics and Automation, pp. 2897–2903, 2009.
[91]	E. Park and S. G. Meek. Fatigue compensation of the electromyographic signal for prosthetic control and force estimation. IEEE transactions on biomedical engineering, 1993, 40(10): 5
[92]	T. V. Camata, J. L. Dantas, T. Abrao, M. A. Brunetto, A. C. Moraes, and L. R. Altimari. Fourier and wavelet spectral analysis of emg signals in supramaximal constant load dynamic exercise. Conf Proc IEEE Eng Med Biol Soc, 2010, 2010: 1364−7
[93]	D. K. Kumar, N. D. Pah, and A. Bradley. Wavelet analysis of surface electromyography to determine muscle fatigue. IEEE Trans Neural Syst Rehabil Eng, 2003, 11(4): 400−6 doi: 10.1109/TNSRE.2003.819901
[94]	P. M. Fitts and M. I. Posner. Human performance.brooks. Cole, Belmont, CA, 1967, 5: 7−16
[95]	Y. Du, W. Jin, W. Wei, Y. Hu, and W. Geng. Surface EMG-based inter-session gesture recognition enhanced by deep domain adaptation. Sensors, 2017, 17(3): 458 doi: 10.3390/s17030458
[96]	P. Tsinganos, B. Cornelis, J. Cornelis, B. Jansen, and A. Skodras, “Improved gesture recognition based on semg signals and tcn, ” in ICASSP 2019-2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE, 2019, pp. 1169–1173.
[97]	U. Cot? e-Allard, C. L. Fall, A. Drouin, A. ′ Campeau-Lecours, C. Gosselin, K. Glette, F. Laviolette, and B. Gosselin. Deep learning for electromyographic hand gesture signal classification using transfer learning. IEEE Transactions on Neural Systems and Rehabilitation Engineering, 2019, 27(4): 760−771 doi: 10.1109/TNSRE.2019.2896269
[98]	U. Cot? e-Allard, E. Campbell, A. Phinyomark, F. ′ Laviolette, B. Gosselin, and E. Scheme. Interpreting deep learning features for myoelectric control: A comparison with handcrafted features. Frontiers in Bioengineering and Biotechnology, 2020, 8: 158 doi: 10.3389/fbioe.2020.00158
[99]	M. Christ, N. Braun, J. Neuffer, and A. W. Kempa-Liehr. Time series feature extraction on basis of scalable hypothesis tests (tsfresh–a python package). Neurocomputing, 2018, 307: 72−77 doi: 10.1016/j.neucom.2018.03.067
[100]	M. Fındık, S¸. Yılmaz, and M. Koseoglu, “Random forest classification of finger- movements using electromyogram (emg) signals, ” in 2020 IEEE Sensors. IEEE, 2020, pp. 1–4.
[101]	F. Zhuang, Z. Qi, K. Duan, D. Xi, Y. Zhu, H. Zhu, H. Xiong, and Q. He, “A comprehensive survey on transfer learning, ” arXiv preprint arXiv: 1911.02685, 2019.
[102]	B. Fernando, A. Habrard, M. Sebban, and T. Tuytelaars, “Unsupervised visual domain adaptation using subspace alignment, ” in Proceedings of the IEEE international conference on computer vision, 2013, pp. 2960–2967
[103]	F. Schiel, A. Hagengruber, J. Vogel, and R. Triebel, “Incremental learning of emg-based control commands using gaussian processes, ” in Conference on Robot Learning 2020, 2020.
[104]	W. Wei, Q. Dai, Y. Wong, Y. Hu, M. Kankanhalli, and W. Geng. Surface-electromyography-based gesture recognition by multi-view deep learning. IEEE Transactions on Biomedical Engineering, 2019, 66(10): 2964−2973 doi: 10.1109/TBME.2019.2899222
[105]	R. Maldonado and S. M. Harabagiu. Active deep learning for the identification of concepts and relations in electroencephalography reports. Journal of biomedical informatics, 2019, 98: 103265 doi: 10.1016/j.jbi.2019.103265
[106]	A. K. Clarke, S. F. Atashzar, A. Del Vecchio, D. Y. Barsakcioglu, S. Muceli, P. Bentley, F. Urh, A. Holobar, and D. Farina, “Deep learning for robust decomposition of high-density surface emg signals, ” IEEE Transactions on Biomedical Engineering, 2020.
[107]	C. J. De Luca, A. Adam, R. Wotiz, L. D. Gilmore, and S. H. Nawab. Decomposition of surface emg signals. Journal of neurophysiology, 2006, 96(3): 1646−1657 doi: 10.1152/jn.00009.2006
[108]	S. Lobov, N. Krilova, I. Kastalskiy, V. Kazantsev, and V. A. Makarov. Latent factors limiting the performance of semg-interfaces. Sensors, 2018, 18(4): 1122 doi: 10.3390/s18041122
[109]	C. Sapsanis, G. Georgoulas, A. Tzes, and D. Lymberopoulos, “Improving emg based classification of basic hand movements using emd, ” in 2013 35th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC). IEEE, 2013, pp. 5754–5757.