Use of forehead bio-signals for controlling an Intelligent Wheelchair

Abstract

This paper presents a novel method to classify human facial movement based on multi-channel forehead bio-signals. Five face movements form three face regions: forehead, eye and jaw are selected and classified in back propagation artificial neural networks (BPANN) by using a combination of transient and steady features from EMG and EOG waveforms. The identified face movements are subsequently employed to generate five control commands for controlling a simulated Intelligent Wheelchair. A human-machine interface (HMI) is designed to map movement patterns into corresponding control commands via a logic control table. The simulation result shows the feasibility and performance of the proposed system, which can be extended into real-world applications as a control interface for disabled and elderly users.