Lower Limb Motion Recognition Method Based on Improved Wavelet Packet Transform and Unscented Kalman Neural Network

Exoskeleton robot is a typical application to assist the motion of lower limbs. To make the lower extremity exoskeleton more flexible, it is necessary to identify various motion intentions of the lower limbs of the human body. Although more sEMG sensors can be used to identify more lower limb motion intention, with the increase in the number of sensors, more and more data need to be processed. In the process of human motion, the collected sEMG signal is easy to be interfered with noise. To improve the practicality of the lower extremity exoskeleton robot, this paper proposed a wavelet packet transform- (WPT-) based sliding window difference average filtering feature extract algorithm and the unscented Kalman neural network (UKFNN) recognition algorithm. We established an sEMG energy feature model, using a sliding window difference average filtering method to suppress noise interference and extracted stable feature values and using UKF filtering to optimize the neural network weights to improve the adaptability and accuracy of the recognition model. In this paper, we collected the sEMG signals of three muscles to identify six lower limb motion intentions. The average accuracy of 94.83% is proposed in this paper. Experiments show that the algorithm improves the accuracy and anti-interference of motion intention recognition of lower limb sEMG signals. The algorithm is superior to the backpropagation neural network (BPNN) recognition algorithm in the lower limb motion intention recognition and proves the effectiveness, novelty, and reliability of the method in this paper.

[1]  Moriba Jah,et al.  Unscented Schmidt–Kalman Filter Algorithm , 2015 .

[2]  Dezhong Yao,et al.  Extracting time-frequency feature of single-channel vastus medialis EMG signals for knee exercise pattern recognition , 2017, PloS one.

[3]  Yuanqing Xia,et al.  Adaptive Robust Unscented Kalman Filter via Fading Factor and Maximum Correntropy Criterion , 2018, Sensors.

[4]  김상호,et al.  EMG 신호 기반 Artificial Neural Network을 이용한 사용자 인식 , 2016 .

[5]  Chen Peng,et al.  Research on EMG segmentation algorithm and walking analysis based on signal envelope and integral electrical signal , 2018, Photonic Network Communications.

[6]  Angkoon Phinyomark,et al.  Feature Extraction and Selection for Myoelectric Control Based on Wearable EMG Sensors , 2018, Sensors.

[7]  Adel Al-Jumaily,et al.  A Framework of Temporal-Spatial Descriptors-Based Feature Extraction for Improved Myoelectric Pattern Recognition , 2017, IEEE Transactions on Neural Systems and Rehabilitation Engineering.

[8]  Yee Mon Aung,et al.  Estimation of Upper Limb Joint Angle Using Surface EMG Signal , 2013 .

[9]  Clodoaldo Ap. M. Lima,et al.  Assessing fractal dimension methods as feature extractors for EMG signal classification , 2014, Eng. Appl. Artif. Intell..

[10]  Wendong Wang,et al.  Real‐time and user‐independent feature classification of forearm using EMG signals , 2018, Journal of the Society for Information Display.

[11]  Toshio Fukuda,et al.  State of the Art: Bipedal Robots for Lower Limb Rehabilitation , 2017 .

[12]  Jin Hu,et al.  iLeg—A Lower Limb Rehabilitation Robot: A Proof of Concept , 2016, IEEE Transactions on Human-Machine Systems.

[13]  Juvenal Rodriguez-Resendiz,et al.  A Study of Movement Classification of the Lower Limb Based on up to 4-EMG Channels , 2019, Electronics.

[14]  Jianda Han,et al.  A State-Space EMG Model for the Estimation of Continuous Joint Movements , 2015, IEEE Transactions on Industrial Electronics.

[15]  Yudong Zhang,et al.  Intelligent Prediction of Human Lower Extremity Joint Moment: An Artificial Neural Network Approach , 2019, IEEE Access.

[16]  Qingsong Ai,et al.  Research on Lower Limb Motion Recognition Based on Fusion of sEMG and Accelerometer Signals , 2017, Symmetry.

[17]  Birger Jettkant,et al.  Effectiveness of an on-body lifting aid (HAL® for care support) to reduce lower back muscle activity during repetitive lifting tasks , 2019, Journal of Clinical Neuroscience.

[18]  Jun Zhao,et al.  A Multimodal Framework Based on Integration of Cortical and Muscular Activities for Decoding Human Intentions About Lower Limb Motions , 2017, IEEE Transactions on Biomedical Circuits and Systems.

[19]  Zhizeng Luo,et al.  Feature-Level Fusion of Surface Electromyography for Activity Monitoring , 2018, Sensors.

[20]  Anke Xue,et al.  Surface Electromyography Based Estimation of Knee Joint Angle by Using Correlation Dimension of Wavelet Coefficient , 2019, IEEE Access.

[21]  Huan Liu,et al.  Local Band Spectral Entropy Based on Wavelet Packet Applied to Surface EMG Signals Analysis , 2016, Entropy.

[22]  Zengguang Hou,et al.  Estimation of Lower Limb Periodic Motions from sEMG Using Least Squares Support Vector Regression , 2014, Neural Processing Letters.

[23]  Wei Li,et al.  sEMG-Based Identification of Hand Motion Commands Using Wavelet Neural Network Combined With Discrete Wavelet Transform , 2016, IEEE Transactions on Industrial Electronics.

[24]  Deok-Hwan Kim,et al.  sEMG Signal-Based Lower Limb Human Motion Detection Using a Top and Slope Feature Extraction Algorithm , 2017, IEEE Signal Processing Letters.

[25]  H. Parsaei,et al.  Comparative Analysis of Wavelet-based Feature Extraction for Intramuscular EMG Signal Decomposition , 2017, Journal of biomedical physics & engineering.

[26]  Xiaodong Zhang,et al.  Surface EMG based continuous estimation of human lower limb joint angles by using deep belief networks , 2018, Biomed. Signal Process. Control..

[27]  Zhen Liu,et al.  Unscented Kalman Filter-Trained Neural Networks for Slip Model Prediction , 2016, PloS one.

[28]  Abdulhamit Subasi,et al.  Comparison of decision tree algorithms for EMG signal classification using DWT , 2015, Biomed. Signal Process. Control..

[29]  Wei-Ping Zhu,et al.  Wavelet Domain Feature Extraction Scheme Based on Dominant Motor Unit Action Potential of EMG Signal for Neuromuscular Disease Classification , 2014, IEEE Transactions on Biomedical Circuits and Systems.

[30]  Pornchai Phukpattaranont,et al.  Probability Density Functions of Stationary Surface EMG Signals in Noisy Environments , 2016, IEEE Transactions on Instrumentation and Measurement.

[31]  Songyuan Zhang,et al.  Real-Time Evaluation of the Signal Processing of sEMG Used in Limb Exoskeleton Rehabilitation System , 2018, Applied bionics and biomechanics.

[32]  Siti Anom Ahmad,et al.  Classification of ankle joint movements based on surface electromyography signals for rehabilitation robot applications , 2015, Medical & Biological Engineering & Computing.

[33]  Pornchai Phukpattaranont,et al.  Evaluation of feature extraction techniques and classifiers for finger movement recognition using surface electromyography signal , 2018, Medical & Biological Engineering & Computing.

[34]  Javier Ruiz-del-Solar,et al.  EMG Signal Filtering Based on Independent Component Analysis and Empirical Mode Decomposition for Estimation of Motor Activation Patterns , 2017 .

[35]  Oluwarotimi Williams Samuel,et al.  A Robust Sparse Representation Based Pattern Recognition Approach for Myoelectric Control , 2018, IEEE Access.

[36]  Qingsong Ai,et al.  A Subject-Specific EMG-Driven Musculoskeletal Model for Applications in Lower-Limb Rehabilitation Robotics , 2016, Int. J. Humanoid Robotics.

[37]  Qiang Meng,et al.  Virtual Rehabilitation Training System Based on Surface EMG Feature Extraction and Analysis , 2019, Journal of Medical Systems.

[38]  Nurhazimah Nazmi,et al.  A Review of Classification Techniques of EMG Signals during Isotonic and Isometric Contractions , 2016, Sensors.

[39]  Hong-Bo Xie,et al.  Stationary Wavelet-based Two-directional Two-dimensional Principal Component Analysis for EMG Signal Classification , 2017 .