基于脑肌电信号的机械臂控制方法与实现

针对脑-机接口目前存在的输入信息源单一、特征识别准确率低、输出控制指令少的问题,本文提出一种基于脑肌电信号的机械臂控制系统。首先对单侧手臂肌电信号和左右手运动想象脑电信号进行同步采集,然后分别进行特征提取和分类识别；并最终将分类模型应用于机械臂的多指令实时控制中。实验结果表明：20名被试者均实现了机械臂的多指令实时控制,且各动作识别准确率平均达到了95%以上。该系统模型丰富了混合脑-机接口的多样性,为脑-机接口在机械臂的控制应用提供了理论依据和实践基础。

Based on EEG and EMG Signals Robotic Control Method and Realization

Aiming at the problem that the current brain-computer interface has a single input information source, low feature recognition accuracy and few output control commands, this paper proposes a robotic arm control system based on EEG and EMG signals. Firstly, the unilateral arm myoelectric EMG and the left and right hand motion imaging EEG are acquired synchronously, and then feature extraction and classification recognition are performed respectively. Finally, the classification model is applied to the real-time control of the robot arm. The experimental results show that all the 20 subjects achieved real-time control of the manipulator, and the accuracy of each action recognition reached more than 85%. The system model enriches the human-computer interaction-mixed brain-computer interface diversity, and provides a theoretical basis and practical basis for the brain-computer interface technology for robotic control.