Natural grasping movement recognition and force estimation using electromyography

Abstract

Electromyography (EMG) generated by human hand movements is usually used to decode different action types with high accuracy. However, the classifications of the gestures rarely consider the impact of force, and the estimation of the grasp force when performing natural grasping movements is so far overlooked. Decoding natural grasping movements and estimating the force generated by the associated movements can help patients to improve the accuracy of prosthesis control. This study mainly focused on two aspects: the classification of four natural grasping movements and the force estimation of these actions. For this purpose, we designed an experimental platform where subjects could perform four common natural grasping movements in daily life, including pinch, palmar, twist, and plug grasp, to complete target profiles. On the one hand, the results showed that, for natural grasping movements with different levels of force (three levels at 20, 50, and 80%), the average accuracy could reach from 91.43 to 97.33% under five classification schemes. On the other hand, the feasibility of force estimation for natural grasping movements was demonstrated. Furthermore, in the process of force estimation, we confirmed that the regression performance about plug grasp was the best, and the average R2 could reach 0.9082. Besides, we found that the regression results were affected by the speed of force application. These findings contribute to the natural control of myoelectric prosthesis and the EMG-based rehabilitation training system, improving the user’s experience and acceptance.

A Surface Electromyography (sEMG) System Applied for Grip Force Monitoring

“…In this case, the sEMG was acquired using a commercial device, Biopac MP150 (Biopac Systems Inc.). The work of [ 29 ] is based on the Delsys system. Both commercial devices are expensive, while the proposed system reduces the system cost by choosing cost-effective components and optimizing the PCB layout.…”

“…Xu and colleagues used Delsys (Delsys Inc., Natick, MA, USA) to collect sEMG and classified four natural grasping actions with an SVM algorithm, achieving an average accuracy rate of 91.43% to 97.33%. They estimated the corresponding force of the actions using a back propagation (BP) neural network, with an average R 2 of 0.9082, which aids in the natural control of myoelectric prosthetics and the application of EMG-based rehabilitation training systems [ 29 ]. Zhao and others proposed a wearable monitoring device for upper limb rehabilitation, integrating ECG and sEMG sensors.…”

“…Both of all were performed based on the ECRL muscle. The Delsys wireless EMG signal acquisition system has been widely used in various studies [29,52,53], and can be considered as a standard. Due to the small size of the ECRL muscle, it is not possible to simultaneously attach two sensors at the same location for simultaneous acquisition (the system proposed in this manuscript as well as the Delsys system).…”

Using High Density EMG to Proportionally Control 3D Model of Human Hand

“…In recent years, the development of EMG-based hand exoskeletons has mostly been focused on grip and grasp gestures [10]. This is because attempts to control multiple fingers through EMG are typically used for hand prosthetics.…”