Home automation control system implementation using SSVEP based brain computer interface

This study postulates Visual Evoked Potential of Electroencephalogram (EEG) signal for home appliance control system to facilitate the integration and control of the household electrical appliances by the handicapped persons suffering from severe motor disabilities. Brain-computer interface (BCI) acts as a favorable communication interface between the mind and computer empowering neural signals to control peripheral devices. Non-Invasive Mind Machine Interface based on the Steady State Visual Evoked Potential (SSVEP) control modality has been implemented and utilized in controlling the home appliances in an offline environment. SSVEP based EEG signal has been acquired from 10 subjects by placing the electrodes in the visual cortex region of human brain. The subject controls the desired appliance by focusing gaze on one of the four flickering LEDs on the SSVEP box having distinct frequencies of 10, 11, 12 and 13 Hz. The features of acquired EEG data have been determined using Minimum Energy. The classification of feature vector has been done by Linear Discriminant Analysis (LDA). The intended control commands are sent to the Arduino for controlling the home appliances. The data from 10 subjects has been used to train the classifier model which has been used in the validation process. The overall accuracy of classifier model has been found to be 84.8%. The data validation accuracy, using data of two subjects, has been indicated as 80.6% and 81.8% respectively.