A shared-control based BCI system: For a robotic arm control

Abstract

Brain-computer interface (BCI) is a technology that can directly translate human thought into machine commands. The Electroencephalogram (EEG) is a kind of brain activity signal that is widely used in brain-computer interfaces, but has a very low signal-to-noise ratio. Therefore, the use of shared control in the EEG-based BCI is an effective way to improve the efficiency of the system. In these systems, the device itself has a certain of autonomy, could able to complete simple control tasks, human takes responsible for high-level decision-making. The two kinds of intelligence works cooperate and play fully use of each part. In this paper, a shared-control scheme is designed for the brain-actuated manipulator. The computer vision module is developed to detect and locate targets so that human only needs to determine which target is desired. The P300-based BCI is designed for target selection. Compared pure BCI-actuated robot arm system, the system use is largely simplified. The online experiment is executed to verify the effectiveness of our system. Four subjects have participated in the experiment, they were instructed to control the robotic arm to grasp the target bottle and translate it to subject's mouth. The results show that the four subjects could complete at least 90% tasks in the ten trials online test. It proved that the method we proposed in this paper is effective.