计算机工程与应用 ›› 2017, Vol. 53 ›› Issue (3): 154-159.DOI: 10.3778/j.issn.1002-8331.1504-0257

• 模式识别与人工智能 • 上一篇    下一篇

基于稳态视觉诱发电位的在线脑机接口研究

董恩增,郭光瑞,陈  超   

  1. 天津理工大学 自动化学院 复杂系统控制理论及应用重点实验室,天津 300384
  • 出版日期:2017-02-01 发布日期:2017-05-11

Research of steady-state visual evoked potential based online brain-computer interface

DONG Enzeng, GUO Guangrui, CHEN Chao   

  1. Complex System Control Theory and Application Key Laboratory, School of Electrical Engineering, Tianjin University of Technology, Tianjin 300384, China
  • Online:2017-02-01 Published:2017-05-11

摘要: 针对脑机接口中存在的抗噪声能力差、操作复杂的问题,利用便携式脑电采集设备Emotiv EPOC以及NAO机器人,搭建了一个抗噪能力较好的稳态视觉诱发在线脑机接口系统。该系统采用典型相关性分析进行稳态视觉诱发电位的频率识别。在线实验中受试者通过Emotiv控制NAO机器人运动,四类任务的准确率达到87.50%。在线实验没有回避周围的噪声,表明该系统具有较好的抗噪能力。

关键词: 稳态视觉诱发电位, 脑机接口, Emotiv EPOC, NAO机器人, 典型相关性分析

Abstract: The most brain-computer interface systems perform badly in resisting the noise and seem complicated for operation. This paper proposes a steady-state visual evoked potential based online brain computer interface system employing the portable EEG collecting device Emotiv EPOC and the NAO robot, which has a good ability of anti-noise. This system utilizes canonical correlation analysis to detect the frequency of steady-state visual evoked potential. In online experiment, the NAO robot is controlled by the EEG signals collected from Emotiv EPOC, the average classification accuracy rate of 4 types of tasks achieves 87. 50%. The online experiment doesn’t suffer from the surrounding noise, which shows the system has a good ability to resist the noise.

Key words: steady-state visual evoked potential, brain computer interface, Emotiv EPOC, NAO robot, canonical correlation analysis