A Brain-Robot Interface by BCI based on Repeated Binary CSP

In this paper, due to the low information transfer rate and low recognition accuracy in Brain-Computer Interface (BCI), a four-class motor imagery Brain-Robot Interface based on Repeated Binary Common Spatial Pattern (RB-CSP) and Support Vector Machine (SVM) is proposed. The control strategy is offline training first, online control next-after users finish learning to control their thinking, the system makes pattern recognition on the collected users´ EEG signals and finally translates them into commands to control the movement of the robot. Experiments indicate that the system is able to extract users´ EEG signal features quickly and correctly, translate them into robot´s control instructions, which can be used to make real-time control on robots effectively.