A Soft Exoskeleton Glove for Hand Bilateral Training via Surface EMG

Abstract

Traditional rigid exoskeletons can be challenging to the comfort of wearers and can have large pressure, which can even alter natural hand motion patterns. In this paper, we propose a low-cost soft exoskeleton glove (SExoG) system driven by surface electromyography (sEMG) signals from non-paretic hand for bilateral training. A customization method of geometrical parameters of soft actuators was presented, and their structure was redesigned. Then, the corresponding pressure values of air-pump to generate different angles of actuators were determined to support four hand motions (extension, rest, spherical grip, and fist). A two-step hybrid model combining the neural network and the state exclusion algorithm was proposed to recognize four hand motions via sEMG signals from the healthy limb. Four subjects were recruited to participate in the experiments. The experimental results show that the pressure values for the four hand motions were about −2, 0, 40, and 70 KPa, and the hybrid model can yield a mean accuracy of 98.7% across four hand motions. It can be concluded that the novel SExoG system can mirror the hand motions of non-paretic hand with good performance.

Recent advances in wearable exoskeletons for human strength augmentation

“…Such exoskeletons, therefore, limit the natural joint movements of hands and even restrain the mobility of the user due to complicated electric tethering and the substantial weight of the exoskeletons. In this regard, recent studies demonstrate hand exoskeletons with minimal and lightweight designs that do not interfere with hand movements [58,92,97,98,100,[106][107][108].…”

“…Although the incorporation of the deep-learning algorithm into the whole system enables the active actuation in accordance with the user's intent, a very few studies incorporated the learning algorithm to make the system much easier to use [92,97,108], in such a way that all the user needs to do is to make a pre-trained movement to for the exoskeleton to augment the strength and control for the movement. One of these previous studies demonstrated the deeplearned, soft exoskeleton for the hand motor function rehabilitation purpose as in figure 8(c).…”

RobHand: A Hand Exoskeleton With Real-Time EMG-Driven Embedded Control. Quantifying Hand Gesture Recognition Delays for Bilateral Rehabilitation

“…Control type Accuracy Delay (2) [53] Point-in-Polygon (PIP) (3 predefined gestures) Accuracy: 0.944 - [57] EMG-driven (neural network to determine force) Force error = 20.7 % - [68] Threshold algorithm -- [69] Linear Bayes classifier (6 predefined gestures) Accuracy (1) : 98.1±4.9% Yes [70] Classification algorithm (6 predefines gestures) Accuracy: 86.38 % Yes [71] EMG-based intent inference method -- [72] Neural network -- [73] Forest classifier (3 predefined gestures) Accuracy: 77.9-85.2 % - [74] Neural Network (4 predefined gestures) Accuracy: 98.7±0.53 - (1) Accuracy for neurologically intact subjects. (2) Whether delay analysis of the system is carried out.…”

“…After a stroke, paretic limbs usually suffer muscle spasticity and weakness, mostly in the first weeks after the episode [59]. Usually, flexors have more MyoArmband (Thalmic Labs) Not specified Inference method no no [72] Trigno wireless EMG system (Delsys) Not specified Neural network no no [73] MyoArmband (Thalmic Labs) Not specified Forest classifier yes no [74] Cyton (OpenBCI) PC Neural network yes no (1) Whether information about accuracy of the EMG control is provided. (2) Whether delay analysis of the system is carried out.…”

Solución integrada de control basado en EMG para la rehabilitación de terapias bilaterales en exoesqueleto de rehabilitación de mano Robhand

“…Trabajos previos relacionados [1], [2], [3], [10], [12], [13], [15], [20] presentaban dos inconvenientes: alto coste del sistema debido a la utilización de sistemas de adquisición de señales EMG comerciales (solo [15] utiliza sistemas de bajo coste, considerados como aquellos que cuesta menos de 150 €) y ejecución del procesamiento de las señales y del control en un ordenador, lo que puede incrementar el tiempo de latencia total del sistema. Teniendo en cuenta lo anteriormente expuesto, el sistema electrónico embebido desarrollado se ha diseñado para reducir el coste y el tiempo de latencia del sistema al máximo.…”