Xu Han1, Ke Lin, Shangkai Gao, Xiaorong Gao. 1. Department of Biomedical Engineering, School of Medicine, Tsinghua University, Beijing 100084, People's Republic of China.
Abstract
OBJECTIVE: Human-robot coordination (HRC) aims to enable human and robot to form a tightly coupled system to accomplish a task. One of its important application prospects is to improve the physical function of the disabled. However, the low level of the coordination between human and robot and the limited potential users still hamper the efficiency of such systems. APPROACH: To deal with such challenges, a novel steady-state visual evoked potential (SSVEP) based human-robot coordinated brain-computer interface (BCI) system was proposed to finish a target capturing task. In this system, the robot, by combining the information obtained during the human's natural interaction with itself to capture a target, could optimize the same object capturing task and yield a better performance automatically. The combination of human dealing with the uncertainty problem and the robot dealing with the complexity problem was the key to the system. Meanwhile, an asynchronous BCI based on SSVEP was used as the system interface, and a novel asynchronous recognition algorithm was used to discriminate the electroencephalogram (EEG) signal. MAIN RESULTS: The results show that the proposed system can lower the fatigue level of the subject and simplify the operation of the system. Meanwhile, the signal recognition accuracy and the efficiency of the system were also improved. SIGNIFICANCE: Under the help of the close and natural coordination relationship design between human and robot, and the asynchronous SSVEP based BCI design which requires no limb movement to control a robot, the users would be provided with an accurate and efficient control experience. Moreover, people with severe motor diseases might potentially benefit from such a system. Also, the proposed methods can be easily integrated into other BCI diagrams, which would ameliorate the predicament of the HRC.
OBJECTIVE:Human-robot coordination (HRC) aims to enable human and robot to form a tightly coupled system to accomplish a task. One of its important application prospects is to improve the physical function of the disabled. However, the low level of the coordination between human and robot and the limited potential users still hamper the efficiency of such systems. APPROACH: To deal with such challenges, a novel steady-state visual evoked potential (SSVEP) based human-robot coordinated brain-computer interface (BCI) system was proposed to finish a target capturing task. In this system, the robot, by combining the information obtained during the human's natural interaction with itself to capture a target, could optimize the same object capturing task and yield a better performance automatically. The combination of human dealing with the uncertainty problem and the robot dealing with the complexity problem was the key to the system. Meanwhile, an asynchronous BCI based on SSVEP was used as the system interface, and a novel asynchronous recognition algorithm was used to discriminate the electroencephalogram (EEG) signal. MAIN RESULTS: The results show that the proposed system can lower the fatigue level of the subject and simplify the operation of the system. Meanwhile, the signal recognition accuracy and the efficiency of the system were also improved. SIGNIFICANCE: Under the help of the close and natural coordination relationship design between human and robot, and the asynchronous SSVEP based BCI design which requires no limb movement to control a robot, the users would be provided with an accurate and efficient control experience. Moreover, people with severe motor diseases might potentially benefit from such a system. Also, the proposed methods can be easily integrated into other BCI diagrams, which would ameliorate the predicament of the HRC.