A brain controlled wheelchair based on common spatial pattern

This paper was proposed for the feature extraction problem in Brain Computer Interface (BCI) which was based on the motor imagery. Common Spatial Pattern (CSP) was used to extract useful features from the Electroencephalograph (EEG) signals. Firstly, a preprocessing step was applied to remove noises. Secondly, CSP was used to analyze with EEG signals. Support Vector Machine (SVM) was investigated to classify motor imagery state. The EEG signals of motor imagery provided by dataset I of 2004 BCI Competition III were used for the validation. The results showed that the algorithm can extract the obvious characteristics efficiently. Finally, the proposed method was used in a wheelchair application. Experimental results showed that the proposed approach was promising for implementing human-computer interaction, especially for EEG-based brain controlled wheelchair.