Offline and online approaches for hybrid brain-machine interfaces based on error-related potentials

• Main Authors: Mauro Nascimben, 馬羅奈斯本
• Other Authors: Fang, Wai-Chi
• Format: Others
• Language: en_US
• Published: 2018
• Online Access: http://ndltd.ncl.edu.tw/handle/2x2cs9

碩士 === 國立交通大學 === 電機資訊國際學程 === 106 === Error Related Potentials could be used in Hybrid Brain Computer Interfaces systems as control mechanism to improve the human-machine interaction. Error Related Potentials are naturally elicited brainwaves when a subject recognizes a mistake. In this thesis Error Related Potentials detection is explored using different techniques and approaches. In an offline classification methodology various classifiers are compared while for online detection recurrent neural network and support vector machine are applied on single-trial basis. For offline classification a spiking neural network based on convolution of spike trains is investigated showing a promising positive outcome in detecting Error Related brainwaves