Control of Newly-Designed Wearable Robotic Hand Exoskeleton Based on Surface Electromyographic Signals

Abstract

The human hand plays a role in a variety of daily activities. This intricate instrument is vulnerable to trauma or neuromuscular disorders. Wearable robotic exoskeletons are an advanced technology with the potential to remarkably promote the recovery of hand function. However, the still face persistent challenges in mechanical and functional integration, with real-time control of the multiactuators in accordance with the motion intentions of the user being a particular sticking point. In this study, we demonstrated a newly-designed wearable robotic hand exoskeleton with multijoints, more degrees of freedom (DOFs), and a larger range of motion (ROM). The exoskeleton hand comprises six linear actuators (two for the thumb and the other four for the fingers) and can realize both independent movements of each digit and coordinative movement involving multiple fingers for grasp and pinch. The kinematic parameters of the hand exoskeleton were analyzed by a motion capture system. The exoskeleton showed higher ROM of the proximal interphalangeal and distal interphalangeal joints compared with the other exoskeletons. Five classifiers including support vector machine (SVM), K-near neighbor (KNN), decision tree (DT), multilayer perceptron (MLP), and multichannel convolutional neural networks (multichannel CNN) were compared for the offline classification. The SVM and KNN had a higher accuracy than the others, reaching up to 99%. For the online classification, three out of the five subjects showed an accuracy of about 80%, and one subject showed an accuracy over 90%. These results suggest that the new wearable exoskeleton could facilitate hand rehabilitation for a larger ROM and higher dexterity and could be controlled according to the motion intention of the subjects.

Keywords: exoskeleton; gesture recognition; hand rehabilitation; surface electromyography; wearable robots.

Figure 1

The mechanical design and realization of hand exoskeleton. (A) The mechanical design of the exoskeleton; (B) mechanical structure of exoskeleton index finger; (C) prototype of the hand exoskeleton.

Figure 2

Schema of the control system for the hand exoskeleton based on non-paretic sEMG processing and offline-to-online classifications.

Figure 3

The reflective marker sets for kinematic analysis of the hand exoskeleton. (A) The top view of the marker sets; (B) the profile view of the marker sets; (C) the marker sets for the human hand.

Figure 4

The muscles selection and hand gesture recognition. (A) The eight muscles of the bilateral forearms and hands; (B) the gestures for classifier selection; (C) the gestures for real-time control of exoskeleton.

Figure 5

The multi-channel CNN algorithm. Conv1, convolution layer 1; Conv2, convolution Layer 2; FC1, fully connected layer 1; FC2, fully connected layer 2.

Figure 6

The joint angles and trajectories of the exoskeleton and human digits during flexion and extension. (A) The joint angles of the metacarpophalangeal joint (MCP), PIP, and DIP of the exoskeleton index finger; (B) the joint angles of the MCP, PIP, and DIP of the index finger of a representative subject; (C) the trajectories of the exoskeleton fingertips.

Figure 7

The co-contraction index (CI) matrices for the four gestures of the left and right hands.

Figure 8

The online and offline classification accuracies. (A) The accuracies of the offline classification for classifier selection; (B) the accuracies of the online and offline classifications for real-time control of exoskeleton using the support vector machine (SVM).

Figure 9

The subject-specific online classification accuracies for recognizing the hand gestures.

Figure 10

Real-time hand gesture recognition from a representative subject (H4). (A) The raw surface electromyography (sEMG) signals recorded from the left brachioradialis (BRA); (B) the Teager Kaiser energy (TKE) signals extracted from (A) with a threshold 0.001 mv; (C) the raw classification based on the SVM; (D) final classification based on the TKE and the three consecutive judgment algorithms.