Use this URL to cite or link to this record in EThOS:
Title: Face movement based multi-modality human machine interface for intelligent wheelchairs
Author: Wei, Lai
ISNI:       0000 0004 2738 2176
Awarding Body: University of Essex
Current Institution: University of Essex
Date of Award: 2012
Availability of Full Text:
Access from EThOS:
This thesis is dedicated to the development of advanced human machine interfaces (HMIs) for elderly people and people with motor disabilities such as spinal cord injury, quadriplegia and amputation. Based on an intelligent wheelchair system developed at the University of Essex, several new face movement based control interfaces are unveiled and utilized to substitute traditional joystick controllers. Color face video from a web camera is applied to track face, head and eye movements of the user. Meanwhile, the concept of multi-modality human machine interface (MMHMI) is introduced to integrate face video data with facial muscle and eye activities recorded from electromyography (EMG) and electrooculography (EOG) signals. The control interface presented in this thesis covers the state-of-the-art research progress carried out under the intelligent wheelchair project at the robot arena for the last five years. In order to test the mechanism between different wheelchair control methods, three face movement based control interfaces are deployed in indoor navigation tasks in which users are asked to drive a wheelchair following designated routes, avoid obstacle and barriers one by one. The control results are compared with traditional joystick and touch screen control methods. The performance of individual control methods is analyzed based on recorded user behavior, wheelchair trajectory, task completion duration,' and user feedback. User driving experience is collected in the form of feedback questionnaires in terms of the wheelchair control such as user comfort, interface response, control accuracy and control reliability. These can be are used to improve future wheelchair control interface design.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available