The Research Of Brain Controlled Drawing Machine Based On SSVEP

Brain-computer Interface (BCI) is an intelligent system which can translate the intentional EEG to the control commands. It is different from the common way that the human being communicates with each other by utilizing muscle tissue. It realizes that the brain can directly communicate with an external device and constructs a new form of exporting brain information. It has broad applicable prospects in the medical recovery、 intelligent control、entertainment and other fields.At present, the development of BCI is gradually developing from the laboratory research stage to the practical application stage, particularly the BCI which is based on steady-state visual evoked potential (SSVEP) has special advantages such as simple operation, the high information translate rate, without training, etc. That makes it become popular in the brain research field. During the BCI experiments which are based on the steady-state visual evoked potential, the eyes of the operator are suffered from long and repeated stimulation of visual stimulator. Visual fatigue is a problem that needs to be solved.This paper had designed and implemented a brain control drawing machine which is based on SSVEP, the system can help the paralyzed people realize their drawing dream. Mainly research works is as follows:1. It elaborated common analytical method of SSVEP signal, this paper employed canonical correlation analysis (CCA) and the power spectrum estimation (PSDA) to fast analyse the SSVEP signal, and compared the results of two methods in feature extraction from SSVEP signal. Synthetic results show that the CCA method can more accurately extract SSVEP signal frequency characteristics and improve system all performance.2. This paper designed two stages of steady-state visual evoked potentials experiment to study detecting EEG fatigue based on SSVEP BCI. In this paper, we employed power spectrum estimation methods to compute the EEG band spectrum. By comparing the EEG band spectral index of the Alert stage and fatigue stage, combined with subjective fatigue scale to comprehensively compute subjects fatigue level. According to the result, we improved the visual stimulator in order to relieve the subject’s visual fatigue and develop the performance of system.3. This paper had been designed and implemented a brain controlled drawing machine based on SSVEP. We designed the program of a brain control drawing machine based on SSVEP which is realized by using Lab VIEW platform environment. The program can implement multi-frequency visual stimulate module, and employ FIR band-pass filter and power spectrum estimation and other digital signal processing method to extract SSVEP signal frequency feature, matching frequency characteristics with the template feature, judging the target the operator gazed and generating a control command, so it can realize the mechanical arm drawing control of subject. The system achieves a 98% accuracy rate on average multi-tasking, it also indicates the program is feasible.