Design Of An Ssvep-Based Bci System With Vision Assisted Navigation Module For The Cooperative Control Of Multiple Robots

Conventional brain-computer interface (BCI) systems can control a robot to execute a certain task according to human intention. However, in the actual application, it is still difficult for users to control multiple robots to execute the cooperative service tasks, including mobility, manipulation and delivery. At the same time, using BCI to control robots for a long time may bring users inevitable fatigue. To solve these dilemmas, we design an SSVEP-based BCI system for a manipulator and a mobile robot to work cooperatively. To reduce the user's fatigue, a ceiling feature based vision assisted navigation module is developed to facilitate the robot moving process. Considering characteristics of electroencephalograph signals, steady-state visual evoked potentials (SSVEP) are employed as input signals for the BCI system, and seven visual stimuli are modulated as control commands. Consequently, this SSVEP-based BCI system can generate four commands to control the movement of Turtlebot mobile robot, one switch command for the visual assisted navigation module, and two commands for the grasp task of UR5 manipulator. Due to the ceiling feature based vision assisted navigation module, the mobile robot can move effectively without human interference, which may reduce the user's fatigue. To evaluate the performance of this system, three subjects were recruited to participate in the experiments. All the subjects succeeded to execute the cooperative control of multiple robots with the vision assisted navigation module. According to the experimental results, the developed BCI system can effectively realize the cooperative control of multiple robots without much fatigue.