计算机与现代化

• 人工智能 • 上一篇    下一篇

基于SSVEP与运动想象的实时脑控阿凡达系统

  

  1. 广东工业大学自动化学院,广东-广州-510006
  • 收稿日期:2017-05-03 出版日期:2018-01-23 发布日期:2018-01-24
  • 作者简介:林铭铎(1995-),男,广东揭阳人,广东工业大学自动化学院本科生,研究方向:脑电信号处理,机器学习; 欧祖宏(1996-),男,广东广宁人,本科生,研究方向:脑机接口。

Real-time Brain-controlled Avatar System Based on SSVEP and Motor Imagery

  1. School of Automation, Guangdong University of Technology, Guangzhou 510006, China
  • Received:2017-05-03 Online:2018-01-23 Published:2018-01-24

摘要: 针对当前在线脑机接口交互系统识别正确率比较低、模式单一、算法复杂度高等问题,设计基于SSVEP和运动想象的实时脑控实时算法并应用于脑控阿凡达系统。通过对6名受试者进行离线训练和在线测试,实验结果表明该系统较好地反映受试者的控制意图,所有受试者均可以基本准确地控制机器人,可实现多指令的控制,识别准确率高,可进一步推动脑机接口在实际领域的应用和发展。

关键词: 稳态视觉诱发, 运动想象, 脑机接口, 脑控阿凡达

Abstract: Aiming at solving the problem that the accuracy rate of recognition system is relatively low, the model is simple and the complexity of the algorithm is high, the real-time brain control real-time algorithm based on SSVEP and motion imagination is designed and applied to the brain-controlled Avatar system. By conducting off-line training and online testing of six subjects, the experimental results show that the system better reflects the control intention of the subject. All subjects can basically control the robot and achieve multi-instruction control with high recognition accuracy. The system steps to promote the brain-computer interface in the practical application and development.

Key words: motor imagery, brain-computer interface, brain-controlled Avatar