| Brain-computer interface(BCI)can realize direct communication between the human brain and the external environment,and help disabled people to regain the ability to communicate and control.In recent years,steady-state visual evoked potential(SSVEP)BCI has been widely studied and applied because of its simple use,free training,and high information transfer rate(ITR).However,many research results still have a certain gap in usability and effectiveness for practical applications.In view of this problem,this article conducts research on the application of SSVEP-BCI in order to promote the practical process of BCI.In SSVEP-BCI,visual stimulators play an important role.Because computer monitors can present complex visual stimuli and their implementation methods are relatively flexible,SSVEP-BCI which using computer monitors as visual stimulators have attracted much attention in recent years.However,due to the hardware limitations of the display,it is difficult to make optimal choices in portability and multi-command presentation.In order to promote the practical of SSVEP-BCI,this study introduces augmented reality(AR)technology and tries to use it as a visual stimulator.Compared to computer monitors,AR devices are more portable and it is unnecessary for users to switch back and forth between visual stimuli and feedback.Also,Holo Lens glasses were used to implement AR technology,and a system was established to verify the feasibility and effectiveness of AR technology in BCI through following experiments.First,a feasibility study was performed here.Four subjects participated in the experiment,and the feasibility verification and stability test of AR were tested.(1)The effect of AR equipment on EEG signals acquisition was studied to verify the feasibility.Comparative experiments showed that the AR equipment had no impact on the SSVEPBCI system.(2)The stability research was to verify the accuracy of the flicker frequency of the AR equipment.Compared the AR-induced SSVEP with traditional display-induced SSVEP,the accuracy of EEG signal classification in AR environment was 44.27%,83.85%,93.23%,98.44% and 98.44% respectively when the data length was 0.5 s,1.0 s,1.5 s,2.0 s and 2.5 s.The corresponding accuracy rate of the display was 73.44%,95.31%,98.44%,99.48% and 99.48%.It can be seen that the performance of traditional display was better than AR when the data length was less than 2.0 s,but the accuracy of both was similar when the data length was more than 2.0 s.Subsequently,the validity of this study was verified by two experiments.A system for controlling the robotic arm by SSVEP induced by AR was designed.The target object is identified and located by computer vision.Finally,the target control mode was used to control the robotic arm to grasp objects with different characteristics.Under the four-target condition,12 subjects participated in the online experiment.The subject controlled the robotic arm to grasp the object of three different colors and the same shape.The wrong target could be cancelled in time.In the end,the accuracy of the experiment reached 97.71% and ITR reached 16.31 bits / min.At the condition of nine goals,11 subjects participated in the online experiment.Subjects can control the robotic arm to grasp objects of different colors and shapes.ITR reaches 20.57 bits / min.Effectiveness and practicability of the system was finally proved based on the results of above experiments.Therefore,a portable SSVEP-BCI system can be realized in our study. |
PDF Full Text Request