Research And Application Of UAV Control Based On Brain-computer Interface

Brain-Computer Interface(BCI)is a technology that does not rely on normal output channels such as peripheral nerves and muscles,and communicates directly with the outside world through Electroencephalogram(EEG)signals.The technology is now widely used in many fields.By solving the key technical problems of BCI,it will also promote the development and intersection of related scientific fields,thereby inspiring new ideas,exploring new means and exploring new directions.It has the double value of science and application.This paper aims at the practical application of BCI system,and proposes a set of BCI UAV control system based on the combination of Steady-State Visual Evoked Potential(SSVEP)EEG signals and Electrooculogram(EOG)signals.Firstly,the SSVEP stimulation paradigm,blink state detector and brain-computer interface control UAV system experimental process are designed,and its feasibility is verified through offline experiments.Secondly,study the processing method of SSVEP EEG signals.In this paper,wavelet denoising is used to preprocess the EEG signal to obtain an EEG signal with higher signal-to-noise;then an improvement is proposed based on traditional power spectral density analysis(PSDA)and canonical correlation analysis(CCA)Algorithm-hybrid power spectral density analysis and individual signal template-based canonical correlation analysis(Hybrid-PSDA-ISTCCA),this algorithm combines individual signal templates while combining PSDA and CCA algorithms,which effectively improves the average classification accuracy of the system;Blink state detection and idle state detection algorithms improve the practicality of the system.It can be seen from the offline analysis results that the BCI system has effectively improved the average classification accuracy and information transfer rate(ITR),which provides a guarantee for improving the performance of the BCI control UAV system.Thirdly,in order to verify the feasibility of brain-controlled UAVs,a set of BCI-oriented UAV systems is proposed and designed,including hardware design and software design.The hardware part mainly includes the design of the microprocessor module,attitude detection module,power management module,wireless communication module,height measurement module and horizontal fixed-point module;the software part optimizes the control algorithm and control instructions of the UAV according to the characteristics of the EEG signal,realizes the stable flight of the UAV through sensor datafusion,and builds a physical UAV platform based on the above software and hardware structure.Finally,based on the above improved algorithm and UAV platform,a portable brain-controlled UAV system based on SSVEP is designed.The performance of the system is evaluated by using the EEG signal control virtual robot TAIGo and the EEG signal control UAV.In the experiment,6 healthy subjects were selected for on-line control test and data analysis.The experimental results show that the classification accuracy based on Hybrid-PSDA-ISTCCA algorithm is more than 90%,and the average information transfer rate is about 44 bit / min.The experimental results verify the practicability and reliability of the system.