Wheelchair Controlling by eye movements using EOG based Human Machine Interface and Artificial Neural Network (original) (raw)

International Journal of Computer Applications

The use of vital signals as a connection interface between humans and computers has recently attracted a great deal of attention. The electro-oculogram (EOG) signal, which is due to eye potential, is one of these signals. More advanced, EOGbased Human-Machine Interfaces (HMIs) are widely investigated and considered to be a noble interface option for disabled people. Artificial neural networks were utilized in this study to detect eye movement from the EOG signal. Neural networks can detect and classify biological signals with nonlinear dynamics, including EOG signals, due to their ability to learn nonlinear dynamics and their pervasive approximation. In this study, two fundamentally distinct networks, MLP and ART, were used to detect sequential and random eye movements for controlling wheelchair. The results indicate that the MLP network could indeed detect consecutive eye movements with an accuracy of over 90%, although the accuracy of this network detection in the case of random movements is relatively poor. In the field of random eye movements, the greatest results are obtained using the ART2AE network, which allows having a diagnostic accuracy of over 70%.

Sign up for access to the world's latest research.

checkGet notified about relevant papers

checkSave papers to use in your research

checkJoin the discussion with peers

checkTrack your impact