Study And Implementation Of Virtual Reality Interactive System Based On Electroencephalogram And Electrooculogram

As for brain-computer interface(BCI)system,electroencephalogram(EEG)is one of the most popular control signals for it has the advantages of safety,convenience and fast response in control.At the same time,electrooculogram(EOG)is also being applied to a variety of control systems by more and more researchers,for its mode is obvious and easy to detect,and the EOG-based control system will not cause excessive discomfort to the users or make it difficult for operating.With the rise of virtual reality(VR)technology,VR has shown great potential in improving some problems in control systems based on EEG and EOG,which can be used as a signal stimulation and feedback tools that makes the system have better practical performance.Therefore,this dissertation focuses on the virtual reality interactive system based on EEG and EOG,and the main work is as follows:(1)A new P300 stimulation paradigm based on the 3D stereo vision for the virtual reality environment was proposed.This paradigm is based on the feature that users can see 3D stereoscopic scenes in virtual scenes,using the process of virtual objects changing from 2D plane to 3D stereo to construct P300 stimulation signals.Such 3D stereo paradigm contains changes in depth information that are not available in the 2D plane paradigm,and has the potential to induce P300 with better quality.A comparison experiment was designed in a VR environment to compare the performance difference between the P300 system based on the 3D stereo paradigm and the 2D plane paradigm,respectively.The experimental results showed that the P300 system based on the proposed 3D stereo paradigm has better performance in classification accuracy,information transfer rate,and system response time,and the effectiveness of the proposed 3D stereo paradigm was confirmed.(2)A virtual boat driving system based on motor imagery and EOG was designed,which combines motor imagery and EOG to build a hybrid control interface.The motor imagery potential has a fast response and can be adjusted entirely by the users,and this control method can output continuous control signals.On the other hand,the mode of EOG is obvious and easy to detect,and this control method is suitable for the fast output of discrete control signals.The hybrid control interface combines the advantages of both control methods to achieve more efficient,complex,and robust control.A total of 4 healthy volunteers participated in two experiments to evaluate the performance of the system.The results of the experiments and load survey showed that the system can achieve more efficient,complex,and robust control,with a certain degree of practicality.