| Brain computer interface technology can help people with disabilities live better and more independently.Currently,brain computer interfaces based on Steady State Visual Evoked Potential(SSVEP)control external devices using a single algorithm to directly control their movement,resulting in low recognition accuracy,complex operations,and difficulty in completing complex tasks.In order to provide disabled individuals with a better user experience of brain computer interface devices and complete more complex tasks with higher accuracy,this thesis designs a brain controlled automatic path finding car system based on SSVEP.In EEG signal processing,in order to improve the universality and accuracy of SSVEP recognition,this thesis proposes a SSVEP recognition algorithm based on the combination of filter bank canonical correlation analysis(FBCCA)and power spectral density(PSD)analysis.This method uses FBCCA to find high similarity reference frequency signals,and then locks the final response frequency through multiple sets of PSD analysis,Complete frequency identification.This method can achieve high recognition accuracy without training.The experimental results show that when the stimulus duration is 1 second,this algorithm can achieve an accuracy of 86.61%,which is5.44%higher than the PSD analysis method,10.38%higher than the Canonical Correlation Analysis(CCA)method,and 8.86%higher than the FBCCA method.The effectiveness of this algorithm in SSVEP EEG signal recognition and classification has been confirmed.In terms of path planning,in order to achieve faster calculation speed,a Voronoi graph based obstacle avoidance shortest path planning algorithm is proposed.The Voronoi diagram is constructed according to the obstacle location information,and then the Voronoi diagram is expanded to simple polygon,and the shortest Euclidean path within the simple polygon is found to complete the path planning.The algorithm complexity of this algorithm is O(nlogn)for path planning algorithms less than A~*Path planning algorithm and Dijkstra algorithm.In this thesis,the brain controlled automatic road finding vehicle system based on SSVEP is carried out with the brain controlled automatic road finding vehicle control experiment and the Mecanum wheel vehicle obstacle adjustment experiment.The analysis of the experimental results shows that the research carried out in this thesis can process EEG signals well,and control the Mecanum wheel vehicle automatic road finding through classified commands to complete the specified goal.This provides a new approach for the practical application of brain computer interface systems in controlling external devices. |
