Brain Machine Interface Based Teleoperation Control Of Robot Arm

Brain-computer interface(BCI),also known as "brain port" or "brain-computer fusion perception",is an interface that controls the external devices by analyzing EEG signals.As a new type of human-computer interaction,it is becoming a hot topic in neuroscience research,and it has great application potential in psychological cognition,intelligent control and rehabilitation.In this paper,based on the analysis of steady-state visual evoked potentials,a teleoperation control system of manipulator based on brain-computer interface is designed and built.Steady-state visual evoked potentials,when the retina is more than 4Hz visual stimulation,the cerebral cortex occipital region will produce the same frequency of visual stimulation(or multiple)frequency of electrical activity.BCI system based on SSVEP has a high information transfer rate,and the calibration time is short,thus it has been a hot topic in BCI research.In this paper,the brain-computer interface system based on SSVEP is built by analyzing the physiological basis of EEG and designing visual stimulator.The SSVEP signals were ana-lyzed at four different stimulus frequencies,and the recognition results of three different alg-orithms were compared.Finally,the recognition and classification of SSVEP signals based on the multi-channel synchronization index(MSI)identification algorithm was carried out.Combining with the SSVEP-BCI system,this paper designs and implements a robot arm teleoperation system based on brain-computer interface.The result of the EEG recognition will be transmitted to the client through the network and converted into the trajectory of the end-effect of robot arm in the task space to realize tracking control of the manipulator.Real--time monitoring of the remote manipulator is realized by the local server in the visual feedback interaction design of the system.As the communication process in the system need to transfer a large number of images,so this paper uses compression-based image compression to reduce the burden of network communication.At the same time,using the image matching based on the Hausdorff distance,the real-time feedback of robot arm position error is realized,and the control condition of the manipulator is improved.Finally,the system was validated by designing an EEG manipulator to complete the task of reaching the target position.The validity and reliability of the brain-control teleoperation system of the manipulator designed in this paper are validated.