Humanoid Robot Control System Based on AR-SSVEP

In the brain computer interface (BCI), steady-state visual evoked potential (SSVEP) is a relatively common input signal of human-computer interaction systems. However, it often requires a fixed computer screen as a visual stimulator, which limits the flexibility of its application. In this research, HoloLens glasses are used as visual stimulators in a BCI system based on augmented reality to control the humanoid robot NAO to recognize and grasp objects. The system uses augmented reality device to induce steady-state visual evoked potential. The user does not need to perform visual stimulation at a fixed position, which can enhance the applicability in complex environments, thereby achieving more natural human-computer interaction. In order to achieve grasping, this study uses robot monocular vision recognition and establish forward and inverse kinematics models of the robot arms. EEG experiments have been performed to verify the accuracy of the system, it is more flexible and convenient for using augmented reality as stimulators in a humanoid robot control system based on SSVEP-BCI.