Research On Command Recognition Algorithm Of Brain Controlled Wheelchair Based On Steady State Visual Evoked Potential

With the rapid development of artificial intelligence technology,the intelligent system based on Brain Computer Interface(BCI)is becoming more and more perfect.Since steady-state Visual Evoked Potentials(SSVEP)EEG signals have high signal-to-noise ratio,SSVEP-BCI system becomes one of the priority programs in the actual product development.At present,the intelligent system based on SSVEP still has the problems of low instruction recognition rate and low execution success rate in practical application,which limit its practical application.Aiming at the above two problems,this paper studied the command recognition algorithm based on non-invasive SSVEP EEG signals,improved the accuracy of command recognition,and realized the brain control of the wheelchair forward,backward,left,right and stop five states.The major work is following:1.In order to solve the problem of low purity of EEG signal actually collected,Extreme-point Symmetric Mode Decomposition(ESMD)is imported from the routine Empirical Mode Decomposition(EMD)algorithm.By comparing and analyzing the recognition accuracy of SSVEP EEG signals,it is textified that ESMD decomposition has a better effect.2.In order to improve the recognition rate of instruction,Deep Canonical Correlation Analysis(DCCA)is imported from conventional Canonical Correlation Analysis(CCA),by computing the recognition accuracy under different combination algorithms,this paper finally verifies the DCCA algorithm based on ESMD is most superior.The recognition accuracy of this algorithm is 95.6% based on the data set of Tsinghua University and 93.0% based on the data set of laboratory.3.In order to verify the execution effect of the instructions,a brain-controlled wheelchair system was developed,in which the user wore an EEG cap and sent out instructions to control the wheelchair through SSVEP EEG signals.The user could control the wheelchair in five modes of forward,backward,left,right and stop in real time,and the success rate of wheelchair motor command execution reached 98.3%.This paper provides a practical solution for the promotion and application of the BCI control system based on SSVEP EEG signal.