An Exosuit System With Bidirectional Hand Support for Bilateral Assistance Based on Dynamic Gesture Recognition.

Hand motor impairment has seriously affected the daily life of the elderly. We developed an electromyography (EMG) exosuit system with bidirectional hand support for bilateral coordination assistance based on a dynamic gesture recognition model using graph convolutional network (GCN) and long short-term memory network (LSTM). The system included a hardware subsystem and a software subsystem. The hardware subsystem included an exosuit jacket, a backpack module, an EMG recognition module, and a bidirectional support glove. The software subsystem based on the dynamic gesture recognition model was designed to identify dynamic and static gestures by extracting the spatio-temporal features of the patient's EMG signals and to control glove movement. The offline training experiment built the gesture recognition models for each subject and evaluated the feasibility of the recognition model; the online control experiments verified the effectiveness of the exosuit system. The experimental results showed that the proposed model achieve a gesture recognition rate of 96.42% ± 3.26 %, which is higher than the other three traditional recognition models. All subjects successfully completed two daily tasks within a short time and the success rate of bilateral coordination assistance are 88.75% and 86.88%. The exosuit system can effectively help patients by bidirectional hand support strategy for bilateral coordination assistance in daily tasks, and the proposed method can be applied to various limb assistance scenarios.

Development and Characterization of the Shale Stratum Well Wall Stabilized with Nanosomal Sealing Agent.

The development of micro cracks in shale formations can easily lead to wellbore instability caused by liquid phase invasion. In order to effectively seal the shale micropores, the surface treatment of nano-SiO2 particles was developed using the silicane coupling agent A-1891. Then, the temperature-sensitive polypenic acrylamide polymer was modified onto the surface of the nanoprocal particle through reaction to obtain the nanosomal blocking agent ASN. The infrared spectrum shows that there are chemical bonds between the generated polymer chains, rather than simple physical composites, indicating the successful synthesis of the temperature-responsive nanosealing agent ASN. The particle size analysis showed that the synthesized nanoparticles in ASN have a uniform particle size distribution and display no agglomeration phenomenon. Applying ASN as a sealing agent in drilling fluid effectively fills the nanoscale micropores and microcracks in shale, making shale denser and significantly improving the wellbore stability of shale formations. In addition, it has good temperature resistance, can adapt to reservoirs at different temperatures, is non-toxic and environmentally friendly, and has good prospects for stable applications in shale formation wellbore.

An Upper-Limb Rehabilitation Exoskeleton System Controlled by MI Recognition Model With Deep Emphasized Informative Features in a VR Scene.

The prevalence of stroke continues to increase with the global aging. Based on the motor imagery (MI) brain-computer interface (BCI) paradigm and virtual reality (VR) technology, we designed and developed an upper-limb rehabilitation exoskeleton system (VR-ULE) in the VR scenes for stroke patients. The VR-ULE system makes use of the MI electroencephalogram (EEG) recognition model with a convolutional neural network and squeeze-and-excitation (SE) blocks to obtain the patient's motion intentions and control the exoskeleton to move during rehabilitation training movement. Due to the individual differences in EEG, the frequency bands with optimal MI EEG features for each patient are different. Therefore, the weight of different feature channels is learned by combining SE blocks to emphasize the useful information frequency band features. The MI cues in the VR-based virtual scenes can improve the interhemispheric balance and the neuroplasticity of patients. It also makes up for the disadvantages of the current MI-BCIs, such as single usage scenarios, poor individual adaptability, and many interfering factors. We designed the offline training experiment to evaluate the feasibility of the EEG recognition strategy, and designed the online control experiment to verify the effectiveness of the VR-ULE system. The results showed that the MI classification method with MI cues in the VR scenes improved the accuracy of MI classification (86.49% ± 3.02%); all subjects performed two types of rehabilitation training tasks under their own models trained in the offline training experiment, with the highest average completion rates of 86.82% ± 4.66% and 88.48% ± 5.84%. The VR-ULE system can efficiently help stroke patients with hemiplegia complete upper-limb rehabilitation training tasks, and provide the new methods and strategies for BCI-based rehabilitation devices.

Bio-inspired soft pneumatic actuator based on a kresling-like pattern with a rigid skeleton.

INTRODUCTION: Biomimetic soft pneumatic actuators (SPA) with Kresling origami patterns have unique advantages over conventional rigid robots, owing to their adaptability and safety. OBJECTIVES: Inspired by cloning and moving behaviors observed from salps, we proposed an SPA based on a Kresling-like pattern with a rigid skeleton. The elongation and output force were tested, and the effectiveness of the applications with the SPA was evaluated. METHODS: The proposed SPA consists of rigid skeletons and a soft skin. The rigid skeletons are constructed using layers of Kresling-like patterns, while a novel extensible inserting structure is devised to replace the folds found in conventional Kresling patterns. This innovative approach ensures that the SPA exhibits axial contraction/expansion motion without any twisting movement. To mimic the bionic characteristics of swimming and ingesting progress of salps, the proposed SPA can perform an axial contraction motion without twisting and a controllable bending motion based on multi-layered Kresling-like patterns; to mimic the cloning and releasing life phenomena of salps, the number of layers of Kresling-like patterns is changeable by adding or reducing skeleton components according to the practical needs. RESULTS: The experimental elongation results on the SPA with multiple layers of Kresling-like patterns show that the elongation can increase to above 162% by adding layers; the experimental output force results show that the three-layer SPA can provide 6.36 N output force at an air flow rate of 10 L/min, and the output force will continue to increase as the number of layers of Kresling-like pattern increases or the air flow rate increases. Further, we demonstrate the applications of the SPA in soft grippers, scissor grippers, claw grippers and pipe crawlers. CONCLUSION: Our proposed SPA can avoid twisting in the radial contraction motion with high elongation and output force, and provide the practical guidance for bio-inspired soft robotic applications.

Multi-Channel FES Gait Rehabilitation Assistance System Based on Adaptive sEMG Modulation.

Functional electrical stimulation (FES) can be used to stimulate the lower-limb muscles to provide walking assistance to stroke patients. However, the existing surface electromyography (sEMG)-based FES control methods mostly only consider a single muscle with a fixed stimulation intensity and frequency. This study proposes a multi-channel FES gait rehabilitation assistance system based on adaptive myoelectric modulation. The proposed system collects sEMG of the vastus lateralis muscle on the non-affected side to predict the sEMG values of four targeted lower-limb muscles on the affected side using a bidirectional long short-term memory (BILSTM) model. Next, the proposed system modulates the real-time FES output frequency for four targeted muscles based on the predicted sEMG values to provide muscle force compensation. Fifteen healthy subjects were recruited to participate in an offline model-building experiment conducted to evaluate the feasibility of the proposed BILSTM model in predicting the sEMG values. The experimental results showed that the [Formula: see text] value of the best-obtained prediction result reached 0.85 using the BILSTM model, which was significantly higher than that using traditional prediction methods. Moreover, two patients after stroke were recruited in the online assisted-walking experiment to verify the effectiveness of the proposed walking-assistance system. The experimental results showed that the activation of the target muscles of the patients was higher after FES, and the gait movement data were significantly different before and after FES. The proposed system can be effectively applied to walking assistance for stroke patients, and the experimental results can provide new ideas and methods for sEMG-controlled FES rehabilitation applications.

Research on the Recognition of Various Muscle Fatigue States in Resistance Strength Training.

Instantly and accurately identifying the state of dynamic muscle fatigue in resistance training can help fitness trainers to build a more scientific and reasonable training program. By investigating the isokinetic flexion and extension strength training of the knee joint, this paper tried to extract surface electromyogram (sEMG) features and establish recognition models to classify muscle states of the target muscles in the isokinetic strength training of the knee joint. First, an experiment was carried out to collect the sEMG signals of the target muscles. Second, two nonlinear dynamic indexes, wavelet packet entropy (WPE) and power spectrum entropy (PSE), were extracted from the obtained sEMG signals to verify the feasibility of characterizing muscle fatigue. Third, a convolutional neural network (CNN) recognition model was constructed and trained with the obtained sEMG experimental data to enable the extraction and recognition of EMG deep features. Finally, the CNN recognition model was compared with multiple support vector machines (Multi-SVM) and multiple linear discriminant analysis (Multi-LDA). The results showed that the CNN model had a better classification accuracy. The overall recognition accuracy of the CNN model applied to the test data (91.38%) was higher than that of the other two models, which verified that the CNN dynamic fatigue recognition model based on subjective and objective information feedback had better recognition performance. Furthermore, training on a larger dataset could further improve the recognition accuracy of the CNN recognition model.

A BCI painting system using a hybrid control approach based on SSVEP and P300.

Brain-computer interfaces (BCIs) can help people with disabilities to communicate with others, express themselves, and even create art. In this paper, a BCI painting system using a hybrid control approach based on steady-state visual evoked potential (SSVEP) and P300 was developed, which can enable simple painting through brain-controlled painting tools. The BCI painting system is composed of two parts: a hybrid stimulus interface and a hybrid electroencephalogram (EEG) signal processing module. The user selects the menus and tools through the SSVEP and P300 stimulus matrices, respectively, and the paintings are displayed in the canvas area of the hybrid stimulus interface in real time. Twenty subjects participated in this study. An offline training experiment was performed to construct the P300 and SSVEP recognition models for each subject; an online painting experiment, which included a copy-painting task and a free-painting task, was performed to evaluate the BCI painting system. The results of the online painting experiment showed that the average tool selection accuracy (88.92 ± 3.94%) of the BCI painting system using the hybrid stimulus interface was slightly higher than that of the traditional brain painting system based on the P300 stimulus interface; the average information transfer rate (ITR) (74.20 ± 5.28 bpm, 71.80 ± 5.15 bpm) in the copy-painting and free-painting tasks of the BCI painting system was significantly higher than that of the traditional brain painting system. Our BCI painting system can effectively help users express their artistic creativity and improve their painting efficiency, and can provide new methods and new ideas for developing BCI-controlled applications.

Wearable Supernumerary Robotic Limb System Using a Hybrid Control Approach Based on Motor Imagery and Object Detection.

Motor disorder of upper limbs has seriously affected the daily life of the patients with hemiplegia after stroke. We developed a wearable supernumerary robotic limb (SRL) system using a hybrid control approach based on motor imagery (MI) and object detection for upper-limb motion assistance. SRL system included an SRL hardware subsystem and a hybrid control software subsystem. The system obtained the patient's motion intention through MI electroencephalogram (EEG) recognition method based on graph convolutional network (GCN) and gated recurrent unit network (GRU) to control the left and right movements of SRL, and the object detection technology was used together for a quick grasp of target objects to compensate for the disadvantages when using MI EEG alone like fewer control instructions and lower control efficiency. Offline training experiment was designed to obtain subjects' MI recognition models and evaluate the feasibility of the MI EEG recognition method; online control experiment was designed to verify the effectiveness of our wearable SRL system. The results showed that the proposed MI EEG recognition method (GCN+GRU) could effectively improve the MI classification accuracy (90.04% ± 2.36 %) compared with traditional methods; all subjects were able to complete the target object grasping tasks within 23 seconds by controlling the SRL, and the highest average grasping success rate achieved 90.67% in bag grasping task. The SRL system can effectively assist people with upper-limb motor disorder to perform upper-limb tasks in daily life by natural human-robot interaction, and improve their ability of self-help and enhance their confidence of life.

Wellbore Stability through Novel Catechol-Chitosan Biopolymer Encapsulator-Based Drilling Mud.

The problem of wellbore stability has a marked impact on oil and gas exploration and development in the process of drilling. Marine mussel proteins can adhere and encapsulate firmly on deep-water rocks, providing inspiration for solving borehole stability problem and this ability comes from catechol groups. In this paper, a novel biopolymer was synthesized with chitosan and catechol (named "SDGB") by Schiff base-reduction reaction, was developed as an encapsulator in water-based drilling fluids (WBDF). In addition, the chemical enhancing wellbore stability performance of different encapsulators were investigated and compared. The results showed that there were aromatic ring structure, amines, and catechol groups in catechol-chitosan biopolymer molecule. The high shale recovery rate demonstrated its strong shale inhibition performance. The rock treated by catechol-chitosan biopolymer had higher tension shear strength and uniaxial compression strength than others, which indicates that it can effectively strengthen the rock and bind loose minerals in micro-pore and micro-fracture of rock samples. The rheological and filtration property of the WBDF containing catechol-chitosan biopolymer is stable before and after 130 °C/16 h hot rolling, demonstrating its good compatibility with other WBDF agents. Moreover, SDGB could chelate with metal ions, forming a stable covalent bond, which plays an important role in adhesiveness, inhibition, and blockage.

Novel Acrylamide/2-Acrylamide-2-3 Methylpropanesulfonic Acid/Styrene/Maleic Anhydride Polymer-Based CaCO3 Nanoparticles to Improve the Filtration of Water-Based Drilling Fluids at High Temperature.

Filtration loss control under high-temperature conditions is a worldwide issue among water-based drilling fluids (WBDFs). A core-shell high-temperature filter reducer (PAASM-CaCO3) that combines organic macromolecules with inorganic nanomaterials was developed by combining acrylamide (AM), 2-acrylamide-2-methylpropane sulfonic acid (AMPS), styrene (St), and maleic anhydride (MA) as monomers and nano-calcium carbonate (NCC). The molecular structure of PAASM-CaCO3 was characterized. The average molecular weight of the organic part was 6.98 × 105 and the thermal decomposition temperature was about 300 °C. PAASM-CaCO3 had a better high-temperature resistance. The rheological properties and filtration performance of drilling fluids treated with PAASM-CaCO3 were stable before and after aging at 200 °C/16 h, and the effect of filtration control was better than that of commonly used filter reducers. PAASM-CaCO3 improved colloidal stability and mud cake quality at high temperatures.

Effect of velocity and acceleration in joint angle estimation for an EMG-Based upper-limb exoskeleton control.

Most studies on estimating user's joint angles to control upper-limb exoskeleton have focused on using surface electromyogram (sEMG) signals. However, the variations in limb velocity and acceleration can affect the sEMG data and decrease the angle estimation performance in the practical use of the exoskeleton. This paper demonstrated that the variations in elbow angular velocity (EAV) and elbow angular acceleration (EAA) associated with normal use led to a large effect on the elbow joint angle estimation. To minimize this effect, we proposed two methods: (1) collecting sEMG data of multiple EAVs and EAAs as training data and (2) measuring the values of EAV and EAA with a gyroscope. A self-developed upper-limb exoskeleton with pneumatic muscles was used in the online control phase to verify our methods' effectiveness. The predicted elbow angle from the sEMG-angle models which were trained in the offline estimation phase was transferred to control signal of the pneumatic muscles to actuate the exoskeleton to move to the same angle. In the offline estimation phase, the average root mean square error (RMSE) between predicted elbow angle and actual elbow angle was reduced from 22.54° to 10.01° (using method one) and to 6.45° (using method two), respectively; in the online control phase, method two achieved a best control performance (average RMSE = 6.87°). The results showed that using multi-sensor fusion (sEMG sensors and gyroscope) achieved a better estimation performance than using only sEMG sensor, which was helpful to eliminate the velocity and acceleration effect in real-time joint angle estimation for upper-limb exoskeleton control.

The Impact of Load Style Variation on Gait Recognition Based on sEMG Images Using a Convolutional Neural Network.

Surface electromyogram (sEMG) signals are widely employed as a neural control source for lower-limb exoskeletons, in which gait recognition based on sEMG is particularly important. Many scholars have taken measures to improve the accuracy of gait recognition, but several real-time limitations affect its applicability, of which variation in the load styles is obvious. The purposes of this study are to (1) investigate the impact of different load styles on gait recognition; (2) study whether good gait recognition performance can be obtained when a convolutional neural network (CNN) is used to deal with the sEMG image from sparse multichannel sEMG (SMC-sEMG); and (3) explore whether the control system of the lower-limb exoskeleton trained by sEMG from part of the load styles still works efficiently in a real-time environment where multiload styles are required. In addition, we discuss an effective method to improve gait recognition at the levels of the load styles. In our experiment, fifteen able-bodied male graduate students with load (20% of body weight) and using three load styles (SBP = backpack, SCS = cross shoulder, SSS = straight shoulder) were asked to walk uniformly on a treadmill. Each subject performed 50 continuous gait cycles under three speeds (V3 = 3 km/h, V5 = 5 km/h, and V7 = 7 km/h). A CNN was employed to deal with sEMG images from sEMG signals for gait recognition, and back propagation neural networks (BPNNs) and support vector machines (SVMs) were used for comparison by dealing with the same sEMG signal. The results indicated that (1) different load styles had remarkable impact on the gait recognition at three speeds under three load styles (p < 0.001); (2) the performance of gait recognition from the CNN was better than that from the SVM and BPNN at each speed (84.83%, 81.63%, and 83.76% at V3; 93.40%, 88.48%, and 92.36% at V5; and 90.1%, 86.32%, and 85.42% at V7, respectively); and (3) when all the data from three load styles were pooled as testing sets at each speed, more load styles were included in the training set, better performance was obtained, and the statistical analysis suggested that the kinds of load styles included in training set had a significant effect on gait recognition (p = 0.002), from which it can be concluded that the control system of a lower-limb exoskeleton trained by sEMG using only some load styles is not sufficient in a real-time environment.

Pillow Support Model with Partitioned Matching Based on Body Pressure Distribution Matrix.

(1) Objective: Sleep problems have become one of the current serious public health issues. The purpose of this research was to construct an ideal pressure distribution model for head and neck support through research on the partitioned support surface of a pillow in order to guide the development of ergonomic pillows. (2) Methods: Seven typical memory foam pillows were selected as samples, and six subjects were recruited to carry out a body pressure distribution experiment. The average value of the first 10% of the samples in the comfort evaluation was calculated to obtain the relative ideal body pressure distribution matrix. Fuzzy clustering was performed on the ideal matrix to obtain the support surface partition. The ideal body pressure index of each partition was calculated, and a hierarchical analysis of each partition was then performed to determine the pressure sensitivity weight of each partition. Using these approaches, the key ergonomic node coordinates of the partitions of four different groups of people were extracted. The ergonomic node coordinates and the physical characteristics of the material were used to design a pillow prototype. Five subjects were recruited for each of the four groups to repeat the body pressure distribution experiment to evaluate the pillow prototype. (3) Results: An ideal support model with seven partitions, including three partitions in the supine position and four partitions in the lateral position, was constructed. The ideal body pressure distribution matrix and ideal body pressure indicators and pressure sensitivity weights for each partition were provided. The pillow that was designed and manufactured based on this model reproduced the ideal pressure distribution matrix evaluated by various groups of people. (4) Conclusion: The seven-partition ideal support model can effectively describe the head and neck support requirements of supine and lateral positions, which can provide strong support for the development of related products.

Evaluation of a New Rotator Cuff Trainer Based on Oscillating Hydraulic Damping.

In order to provide a convenient way to strengthen the rotator cuff muscles and prevent rotator cuff injury, this study designed an innovative strength trainer specifically for shoulder rotator cuff based on oscillating hydraulic damping. We carried out a myoelectric testing experiment to evaluate the shoulder rotation training effect and compared the results with traditional training equipment to verify the feasibility and validity of the new rotator cuff trainer (RCT). Then, we further explored the influence of different training postures and motion speeds on shoulder rotation training. In the experiment, subjects used three types of equipment (RCT, dumbbells and elastic bands) to perform shoulder rotation training under two movement speeds and two motion postures. The surface electromyography (sEMG) signals of targeted muscles were collected in real time and then further analyzed. The experimental results showed that when using the RCT, the muscle force generation sequence was more aligned with the biomechanical principles of shoulder rotation than using the other two training methods, and the target training muscles had the higher percentage of muscle work. During RCT training, a higher speed of movement (120°/s) led to a higher degree of muscle activation; coronal axis rotation was better for the infraspinatus training, and sagittal axis rotation was better for teres minor training. Based on these results, the RCT was proved to be more effective than traditional training methods. In order to exercise the different muscles of rotator cuff more comprehensively and extensively, different postures should be selected. Furthermore, the movement speed can be appropriately increased within the safe range to improve muscle activation.

A Low-Cost End-to-End sEMG-Based Gait Sub-Phase Recognition System.

As surface electromyogram (sEMG) signals have the ability to detect human movement intention, they are commonly used to be control inputs. However, gait sub-phase classification typically requires monotonous manual labeling process, and commercial sEMG acquisition devices are quite bulky and expensive, thus current sEMG-based gait sub-phase recognition systems are complex and have poor portability. This study presents a low-cost but effective end-to-end sEMG-based gait sub-phase recognition system, which contains a wireless multi-channel signal acquisition device simultaneously collecting sEMG of thigh muscles and plantar pressure signals, and a novel neural network-based sEMG signal classifier combining long-short term memory (LSTM) with multilayer perceptron (MLP). We evaluated the system with subjects walking under five conditions: flat terrain at 5 km/h, flat terrain at 3 km/h, 20 kg backpack at 5 km/h, 20 kg shoulder bag at 5 km/h and 15° slope at 5 km/h. Experimental results show that the proposed method achieved average classification accuracies of 94.10%, 87.25%, 90.71%, 94.02%, and 87.87%, respectively, which were significantly higher than existing recognition methods. Additionally, the proposed system had a good real-time performance with low average inference time in the range of 3.25 ~ 3.31 ms.

Study on the Recognition of Exercise Intensity and Fatigue on Runners Based on Subjective and Objective Information.

A running exhaustion experiment was used to explore the correlations between the time-frequency domain indexes extracted from the surface electromyography (EMG) signals of targeted muscles, heart rate and exercise intensity, and subjective fatigue. The study made further inquiry into the feasibility of reflecting and evaluating the exercise intensity and fatigue effectively during running using physiological indexes，thus providing individualized guidance for running fitness. Twelve healthy men participated in a running exhaustion experiment with an incremental and constant load. The percentage of heart rate reserve (%HRR), mean power frequency (MPF) and root mean square (RMS) from surface EMG (sEMG) signals of the rectus femoris (RF), biceps femoris (BF), tibialis anterior muscle (TA), and the lateral head of gastrocnemius (GAL) were obtained in real-time. The data were processed and analyzed with the rating of perceived exertion (RPE) scale. The experimental results show that the MPF on all the muscles increased with time, but there was no significant correlation between MPF and RPE in both experiments. Additionally, there was no significant correlation between RMS and RPE of GAL and BF, but there was a negative correlation between RMS and RPE of RF. The correlation coefficient was lower in the constant load mode, with the value of only -0.301. The correlation between RMS and RPE of TA was opposite in both experiments. There was a significant linear correlation between %HRR and exercise intensity (r = 0.943). In the experiment, %HRR was significantly correlated with subjective exercise fatigue (r = 0.954). Based on the above results，the MPF and RMS indicators on the four targeted muscles could not conclusively identify fatigue of lower extremities during running. The %HRR could be used to identify exercise intensity and human fatigue during running and could be used as an indicator of recognizing fatigue and exercise intensity in runners.

A Brain-Machine Interface Based on ERD/ERS for an Upper-Limb Exoskeleton Control.

To recognize the user's motion intention, brain-machine interfaces (BMI) usually decode movements from cortical activity to control exoskeletons and neuroprostheses for daily activities. The aim of this paper is to investigate whether self-induced variations of the electroencephalogram (EEG) can be useful as control signals for an upper-limb exoskeleton developed by us. A BMI based on event-related desynchronization/synchronization (ERD/ERS) is proposed. In the decoder-training phase, we investigate the offline classification performance of left versus right hand and left hand versus both feet by using motor execution (ME) or motor imagery (MI). The results indicate that the accuracies of ME sessions are higher than those of MI sessions, and left hand versus both feet paradigm achieves a better classification performance, which would be used in the online-control phase. In the online-control phase, the trained decoder is tested in two scenarios (wearing or without wearing the exoskeleton). The MI and ME sessions wearing the exoskeleton achieve mean classification accuracy of 84.29% ± 2.11% and 87.37% ± 3.06%, respectively. The present study demonstrates that the proposed BMI is effective to control the upper-limb exoskeleton, and provides a practical method by non-invasive EEG signal associated with human natural behavior for clinical applications.

Impact of Load Variation on Joint Angle Estimation From Surface EMG Signals.

Many studies use surface electromyogram (sEMG) signals to estimate the joint angle, for control of upper-limb exoskeletons and prostheses. However, several practical factors still affect its clinical applicability. One of these factors is the load variation during daily use. This paper demonstrates that the load variation can have a substantial impact on performance of elbow angle estimation. This impact leads an increase in mean RMSE (Root-Mean-Square Error) from 7.86 ° to 20.44 ° in our experimental test. Therefore, we propose three methods to address this issue: 1) pooling the training data from all loads together to form the pooled training data for the training model; 2) adding the measured load value (force sensor) as an additional input; and 3) developing a two-step hybrid estimation approach based on load and sEMG. Experiments are conducted with five subjects to investigate the feasibility of the proposed three methods. The results show that the mean RMSE is reduced from 20.44 ° to 13.54 ° using method one, 10.47 ° using method two, and 8.48 ° using method three, respectively. Our study indicates that 1) the proposed methods can improve performance and stability on joint angle estimation and 2) sensor fusion (sEMG sensor and force sensor) is an efficient way to resolve the adverse effect of load variation.

An upper-limb power-assist exoskeleton using proportional myoelectric control.

We developed an upper-limb power-assist exoskeleton actuated by pneumatic muscles. The exoskeleton included two metal links: a nylon joint, four size-adjustable carbon fiber bracers, a potentiometer and two pneumatic muscles. The proportional myoelectric control method was proposed to control the exoskeleton according to the user's motion intention in real time. With the feature extraction procedure and the classification (back-propagation neural network), an electromyogram (EMG)-angle model was constructed to be used for pattern recognition. Six healthy subjects performed elbow flexion-extension movements under four experimental conditions: (1) holding a 1-kg load, wearing the exoskeleton, but with no actuation and for different periods (2-s, 4-s and 8-s periods); (2) holding a 1-kg load, without wearing the exoskeleton, for a fixed period; (3) holding a 1-kg load, wearing the exoskeleton, but with no actuation, for a fixed period; (4) holding a 1-kg load, wearing the exoskeleton under proportional myoelectric control, for a fixed period. The EMG signals of the biceps brachii, the brachioradialis, the triceps brachii and the anconeus and the angle of the elbow were collected. The control scheme's reliability and power-assist effectiveness were evaluated in the experiments. The results indicated that the exoskeleton could be controlled by the user's motion intention in real time and that it was useful for augmenting arm performance with neurological signal control, which could be applied to assist in elbow rehabilitation after neurological injury.