Design And Implementation Of Obstacle Avoidance Car Based On SSVEP

The brain-computer interface is a system in which human consciousness is directly connected to the external environment.Steady-state visual evoked potential refers to an EEG signal with a stimulating frequency characteristic obtained by gazing at a certain frequency of stimulation.The SSVEP-based BCI system has been extensively studied in BCI systems because of its simple operation,low training,and high transmission rate.In view of the shortcomings of the current EEG signal acquisition system,which is bulky and not flexible enough,a portable acquisition system is designed to collect EEG signals in real time,and the result of classification is used to control the smart car in an obstacle environment.Firstly,this thesis studies the characteristics and principles of the visual stimulation interface,as well as various influencing factors,and uses the Java language to design a visual stimulation interface based on steady-state visual evoked potential on the LCD.A portable EEG signal acquisition device designed by ADS1299 was used to collect EEG signals of subjects.The EEG signal is filtered to remove noise and amplification,and then the analog-todigital conversion is sent to the host computer through Bluetooth to display it.The portable device is used to successfully collect the EEG signal of the subject and display it on the computer at a higher resolution.According to the two design paradigms of the brain-computer interface: imaginary movement and SSVEP,For the design paradigm of imaginary motion,the algorithm of EMD combined with PSD and CSP is proposed,and the CCA algorithm is proposed for the design paradigm of SSVEP.The signal processing algorithms proposed in both paradigms have higher classification accuracy.Finally,A portable BCI system is designed by using OpenBCI development platform,and the system is used to control the movement direction of the intelligent car in an obstacle environment.In the online system experiment,the average accuracy of the four subjects was close to 90%,and the performance was greatly improved compared with other brain-computer interface systems.In addition,the BCI system of the brain control car designed in this thesis increases the feedback mechanism to reduce the reaction time of the subject and improve the performance of the system.At the same time,the experimental results also verify the real-time effectiveness of CCA algorithm for SSVEP signal feature extraction,and provide a new solution for the application of brain-computer interface system in real life.