Search

Traditional myoelectric prostheses that employ a static pattern recognition model to identify human movement intention from surface electromyography (sEMG) signals hardly adapt to the changes in the sEMG characteristics caused by interferences from daily activities, which hinders the clinical applications of such prostheses. In this paper, we focus on methods to reduce or eliminate the impacts of three types of daily interferences on myoelectric pattern recognition (MPR), i.e., outlier motion, muscle fatigue, and electrode doffing/donning. We constructed an adaptive incremental hybrid classifier (AIHC) by combining one-class support vector data description and multi-class linear discriminant analysis in conjunction with two specific update schemes. We developed an AIHC-based MPR strategy to improve the robustness of MPR against the three interferences. Extensive experiments on hand-motion recognition were conducted to demonstrate the performance of the proposed method. Experimental results show that the AIHC has significant advantages over non-adaptive classifiers under various interferences, with improvements in the classification accuracy ranging from 7.1% to 39% ( ${p} ). The additional evaluations on data deviations demonstrate that the AIHC can accommodate large-scale changes in the sEMG characteristics, revealing the potential of the AIHC-based MPR strategy in the development of clinical myoelectric prostheses.

Skeleton-based human action recognition has recently drawn a lot of attentions with the increasing availability of large-scale skeleton datasets. Graph Convolutional Network (GCN) methods have achieved relatively good performances in action recognition. However, most GCN methods based on predefined graphs with fixed topology constraints always neglect the potential dependencies derived from the cooperative movement of all joints. Besides, the lengths and the directions of skeletons are rarely involved. These easily cause a larger deviation of the estimated action from the actual action. Here, a two-steam fully connected graph convolutional network (2s-FGCN) is proposed. The topology structure of the 2s-FGCN covers the local physical connections and the global potential cooperation of all joints and the joints, lengths and directions of skeletons are all input to the model. The experimental results on two datasets (NTU-RGB+D and Kinetics-Skeleton) demonstrate that the proposed model can obtain the state-of-the-art results.

Due to the couplings among joint-relative muscles, it is a challenge to accurately estimate continuous multi-joint movements from multi-channel sEMG signals. Traditional approaches always build a nonlinear regression model, such as artificial neural network, to predict the multi-joint movement variables using sEMG as inputs. However, the redundant sEMG-data are always not distinguished; the prediction errors cannot be evaluated and corrected online as well. In this work, a correlation-based redundancy-segmentation method is proposed to segment the sEMG-vector including redundancy into irredundant and redundant subvectors. Then, a general state-space framework is developed to build the motion model by regarding the irredundant subvector as input and the redundant one as measurement output. With the built state-space motion model, a closed-loop prediction-correction algorithm, i.e., the unscented Kalman filter (UKF), can be employed to estimate the multi-joint angles from sEMG, where the redundant sEMG-data are used to reject model uncertainties. After having fully employed the redundancy, the proposed method can provide accurate and smooth estimation results. Comprehensive experiments are conducted on the multi-joint movements of the upper limb. The maximum RMSE of the estimations obtained by the proposed method is 0.16±0.03, which is significantly less than 0.25±0.06 and 0.27±0.07 (p < 0.05) obtained by common neural networks.

Decomposition of EMG (electromyography), in which the MUAPs (Motor Units Action Potentials) are extracted from the original EMG signals, is a useful approach for evaluating the physiological properties of muscles and studying the neural mechanism of human motions. Existing decomposition methods usually extracted the MUAPs via the waveform matching between the detected spikes and MUAP templates. However, those methods always involve too many artificial parameters and have no generally mathematical expression, which make them inflexible in practical applications and also cause poor robustness under interference. With respect to the problem, in this paper, a wavelet-based decomposition method is proposed to detect the “true” MUAPs from sEMG. Two standard Gaussian wavelets are firstly defined as the basic MUAP expressions; then, wavelet transform is performed to describe scales and magnitudes of different MUAPs from different nerve fibers. Experiments were conducted to verify the performance of the proposed method, where MUAPs decomposed from sEMG under different levels of muscle contraction can be effectively detected by the method.

Traditional EMG-motion recognition methods are only able to recognize target motions that presented in the training phase, but cannot detect randomly-occurred outlier motions that did not present before. Here, a hybrid classifier that combines one-class SVMs and a multi-class LDA was proposed to perform recognition on target classes and rejection on outlier classes. The classification ability of the hybrid classifier can incrementally grow via online learning the data of outlier classes. Extensive experiments on EMG-based hand-motion recognition were conducted to verify the performance of the incremental hybrid classifier (IHC). The mean recognition accuracy on target classes of IHC is 92%, which is 23% higher than that of the normal MLP. Moreover, IHC has the ability to detect outlier patterns that MLP would misclassify to target classes.

Missing data are a common drawback that pattern recognition techniques need to handle when solving real-life classification tasks. This paper first discusses problems in handling high-dimensional samples with missing values by the Gaussian mixture model (GMM). Since fitting the GMM by directly using high-dimensional samples as inputs is difficult due to the convergence and stability issues, a novel method is proposed to build the high-dimensional GMM by extending a reduced-dimensional GMM to the full-dimensional space. Based on the extended full-dimensional GMM, two approaches, namely, marginalization and conditional-mean imputation, are proposed to classify samples with missing data in online phase. Then, the proposed methods were employed to recognize hand motions from surface electromyography (sEMG) signals, and more than 75% of classification accuracy of motions can be obtained even if 50% of sEMG signals were missing. Comparisons with normal mean and zero imputations also demonstrate the improvements of the proposed methods. Finally, a control scheme for a myoelectric hand was designed by involving the novel methods, and online experiments confirm the ability of the proposed methods to improve the safety and stability of practical systems.

A state-space electromyography (EMG) model is developed for continuous motion estimation of human limb in this paper. While the general Hill-based muscle model (HMM) estimates only joint torque from EMG signals in an “open-loop” form, we integrate the forward dynamics of human joint movement into the HMM, and such an extended HMM can be used to estimate the joint motion states directly. EMG features are developed to construct measurement equations for the extended HMM to form a state-space model. With the state-space HMM, a normal closed-loop prediction–correction approach such as the Kalman-type algorithm can be used to estimate the continuous joint movement from EMG signals, where the measurement equation is used to reject model uncertainties and external disturbances. Moreover, we propose a new normalization approach for EMG signals for the purpose of rejecting the dependence of the motion estimation on varying external loads. Comprehensive experiments are conducted on the human elbow joint, and the improvements of the proposed methods are verified by the comparison of the EMG-based estimation and the inertial measurement unit measurements.

Acupuncture is a traditional Chinese therapeutic method, recognized by western medicine as an important complementary therapy. However, the efficacy assessment of this traditional intervention has been mainly empirical rather than scientific. In this paper, we propose a surface electromyography (sEMG) based approach for quantitative assessment of acupuncture on Bell's palsy. The assessment is made by comparing the muscle activities of healthy with diseased ones. A feature-vector of four individual sEMG features is introduced for more comprehensive representation of muscle activity. A clustering technique is then proposed to calculate quantitative differences between the muscle activities of the healthy and diseased sides. A regressive model is also proposed to predict the recovery trend of a patient based on the quantitative assessments during his/her clinical acupuncture history. As reported in the paper, a total of 20 selected Bell's palsy patients have participated in the extensive experiments that were conducted to verify the performance of the proposed method.

In order to evaluate the kinematic performance and determine the proper design parameters, the kinematic characteristics of human upper-limb are modelled and analyzed in this paper. Based on the model, two different mechanical configurations of wearable upper-limb exoskeleton robots are proposed and the singularity orientation in the upper limb workspace is analyzed. The optimal configuration of the shoulder mechanism without limiting the spherical-joint motion is determined to avoid the singularity in the control procedure. According to the configuration, the workspaces of the upper limb and the exoskeleton robots are simulated by the Matlab software. And the results prove that the optimized configuration works well in the most frequently used workspace of human upper-limb. The optimizing method of singularity elimination is helpful for the configuration design and active control of upper limb exoskeleton system.

Electromyography (EMG) has been popularly used as interface command to achieve a natural control for myoelectric prosthetic-hands. Traditional EMG-based recognition methods always only focus on the classification of target motion classes that were defined in the training phase, but have no ability to reject outlier motion interferences that did not present before. In this paper, a hybrid classifier that combines one one-class Gaussian classifiers and a multi-class LDA was constructed to achieve EMG-based motion classification, in which Gaussian classifiers were used to reject outlier interferences, while LDA was used to classify target motion samples. The robust hybrid classifier is easily built and has low run-time complexity. Extensive experiments were conducted to verify the performance of the proposed hybrid classifier, where 91.6% of target motion recognition accuracy and 96.5% of outlier motion rejection accuracy were respectively obtained. Finally, the hybrid classifier was involved to achieve a robust and real-time control of a myoelectric prosthetic-hand.

sEMG recognition has been used extensively in prosthetic device control, human-assisting manipulators and sign language recognition, etc. However, the sEMG recognition model, trained with one subject's sEMG data, is not applicable to the other subjects, which hinders the practical application of myoelectric interfaces immensely. In this paper, a sEMG recognition method which is applicable to multi-users is proposed. Firstly, single channel sEMG is decomposed into 30 MUAPTs, which includes four steps: two-order differential filter, threshold calculation, spike detection and hierarchical clustering. Secondly, the MUAPTs are updated with the templates orthogonalization; and Deep Boltzman Machine is employed to classify the MUAPTs into five classes corresponding to the predefined five gestures. Six participants participated in this experiment to validate the effectiveness of the proposed method. Results indicated that this method can achieve a mean accuracy of 81.5%.

A novel closed-loop framework for multidimensional sequence processing is proposed in this paper. Traditional sequence-driven models are always forward, so no information is feedback to correct their outputs, which may deviate from the true values gradually due to the estimation error accumulating. To overcome the problem, the multidimensional vector in the input sequence is divided into two vectors based on its data attribute. One vector sequence generated from the original input sequence is considered as the new input sequence, and the other is considered as the measurement output sequence. The original output sequence is treated as the state sequence. Then, a closed-loop model in the state-space form is constructed, with which the states can be estimated online by filtering algorithms. The feasibility of the proposed framework has been verified by using the robot inverse kinematics.

The classic Bays filters applied to model-based visual tracking suffers from high computation complexity and performance degradation when the inaccurate priori knowledge is involved. In order to improve tracking real-time and accuracy, two kinds of adaptive unscented Kalman filters (AUKFs), named the MIT-based AUKF and the master-slave-structure AUKF, respectively, are proposed to estimate the 3-D rigid-body motion from sequential images. The filters use certain feature points' image coordinates as input data to estimate the position and orientation of the object at each instant when an image is captured, and to recover the velocity and angular velocity of the object between consecutive frames. Experimental results show that both the AUKFs can improve estimation real-time and accuracy in visual tracking.

Electromyography (EMG) has been widely used as control commands for prosthesis, powered exoskeletons and rehabilitative robots. In this paper, an EMG-driven state space model is developed to estimate continuous joint angular displacement and velocity, demonstrated by elbow flexion/ extension. The model combines the Hill-based muscle model with the forward dynamics of joint movement, in which kinematic variables are expressed as a function of neural activation levels. EMG features including integral of absolute value and waveform length are then extracted, and two quadratic equations which associate the kinematic variables with EMG features are constructed to represent the measurement equation. The proposed model are verified by extensively experiments, where the angular movements of human elbow joint are estimated only using the EMG signals, and the estimations are compared with the IMU measurements to validate the accuracy. As a demonstration, a robotic arm is commanded to follow the human elbow movement estimated by the proposed model, which shows the possibility of EMG-based robotic assisted rehabilitation.

Predicting Torque by using Electromyography (EMG) signals is a significant problem as long as wearable robots is concerned. In this paper, a coarse-to-fine algorithm, named as Hierarchical Projected Regression, is proposed to estimate the torque of elbow by the surface EMG signals from human's arm. HPR is learned based on a clinical dataset. First, the real torque distribution can be clustered by k-means algorithm. Meanwhile, the EMG signals are divided into multi-clusters corresponding to the partitions of torque dataset. Then, a projected matrix is derived based on the criterion of Linear Discriminant Analysis, by which the low dimensional features are extracted from the original high dimensional EMG. Finally, the process extends to multi-level tree algorithm based on the hierarchical variance reduction of individual torque cluster. The experimental results show the high accuracy and efficiency of this novel method.

Electromyography (EMG) has been widely used as control commands for prosthesis, powered exoskeletons and rehabilitative robots. In this paper, an EMG-driven state-space model is developed to estimate joint angular velocities and angles throughout elbow flexion/extension. The state equation of the model combines the Hill-based muscle model with the forward dynamics of joint movement, and expresses the kinematic variables as a function of neural activation levels. Then, EMG features including integral of absolute value and waveform length are extracted, and two quadratic equations which associate the kinematic variables with EMG features are fitted to represent the measurement equation. Based on the proposed model, the joint angular velocities and angles are estimated just using the EMG signals with the Extended Kalman Filter (EKF), and the estimation results are used to control a manipulator. The experimental results demonstrate the efficiency of EMG-based motion control with the proposed model.