Summary:
Due to the rapid development of technology, the Human Brain and Computers are interfered with by Bio-Electronic devices employing bio-signals, which are detected by a particular class of sensors called bio-sensors. A new emerging research, the study of bio-signals has focused particularly on mind-controlled technology. More specifically, directly controlling a vehicle using brain waves might assist people with impairments regain their driving abilities as well as offer a fresh option for healthy people to operate a vehicle. The current thesis describes a Brain Controlled Vehicle (BCV) that uses Brain Computer Interface (BCI) technology to interpret Electroencephalography (EEG) data, operate a device, and evaluate brain waves, in order to stay as close as possible to the human nature. The system, which is based on Machine Learning techniques, comprises the following features: (a) Processing of EEG data in order to perform various feature extraction methods; (b) make use of a proper dimensionality reduction method that will find correlations in the data and discard non-critical information; (c) implement classification methods that are able to predict the desired motion related labels (left hand, right hand, both feet, tongue); (d) map the predicted motion related labels into real motions (turn left, turn right, accelerate, slow down) and (e) integrate the best models, with the use of a voting method, into a final BCV system.
Keywords:
Brain Controlled Vehicle (BCV), Electroencephalography (EEG), Classification Algorithms, Feature Extraction Methods
