EMG intention perception approach in multimodal human computer interaction of wheelchairs.

Abstract

At present, the limited mobility of the rehabilitation robots hampers stroke recovery. To address this issue, this paper presents multimodal human computer interaction (HCI) of wheelchairs, concentrating on electromyography (EMG) intention perception approach utilized in wheelchair control. In our study, five healthy subjects recorded EMG signals from four gestures. Through mean absolute value (MAV), root mean square (RMS), and probabilistic neural network (PNN) for classification, the EMG intention perception approach successfully facilitated multi-command control of wheelchairs. Although preliminary, this validation shows great promise for revolutionizing assistive mobility, enhancing independence, and significantly improving the quality of life for stroke patients.