A Brain-Vehicle Interaction Method Based On ERD/ERS And Alpha Wave

The thesis is a part of a project entitled “Brain-Computer interaction and control methods of brain-controlled vehicles”,supported by the National Natural Science Foundation under Grant 61374192.A brain-controlled vehicle is a vehicle that directly controlled by the brain without passing through the limbs of the driver.The research about brain-controlled vehicles is of great significance in practical application.It can provide a new channel for the people with disabilities to control a vehicle,which can expand their scope of activities.In addition,it can assist normal people in driving a vehicle to improve the driving performance.This thesis focused on the study of the brain-computer interaction method of the braincontrolled vehicle.In order to improve the classification accuracy of ERD/ERS in two kinds of tasks,we proposed two improved methods,namely,Bandpower-based method and SampEn-based method,and verified the classification performance of the two methods.Based on the method of Bandpower,a brain-computer interface model based on ERD/ERS was established.Further integration of the brain-computer interface model based on alpha wave,we proposed a new brain-computer interaction system,which was the brain-vehicle interaction system based on ERD/ERS and alpha wave,and tested the performance of the system.This thesis achieved the following research accomplishments:1.On the basis of the original ERD/ERS signal recognition method,two new recognition methods were proposed: Bandpower-based method and SampEn-based method.By making use of EEG data in the existing standard library,the recognition performance of the two methods on ERD/ERS was verified,which was better than that of the original method.2.Based on the Bandpower method,the brain-computer interface model based on ERD/ERS was established,and the classification performance of the model was verified.Established brain-computer interface model based on alpha wave,and verified the validity of the model.Based on the above two kinds of brain-computer interface model,a brainvehicle interaction system based on ERD/ERS and alpha wave was constructed.3.Built the experiment platform and tested the performance of the proposed brain-vehicle interaction system by analyzing the success rate of the completion of the task,the time of the completion of the task,the average lateral deviation and other basic indicators.