织物电极在多功能肌电假肢控制中的性能分析PerformanceAnalysis.PDFVIP

Vol. 4 No.11/ Nov. 2010 织物电极在多功能肌电假肢 控制中的性能分析 Performance Analysis of Textile Sensors in Control of Multifunctional Myoelectric Prostheses 张志勇 陶丹丹 耿艳娟 李光林 摘 要 传统肌电假肢控制是通过银/氯化银凝胶金属电极采集肌肉表面电信号(Electromyography, EMG)。尽管导电凝 胶金属电极可以提供高质量的肌电信号，但是它不方便应用于肌电假肢的临床应用。织物电极作为新型的传感器，将会是 多功能肌电假肢控制的信号采集的新尝试。然而，用织物电极采集的肌电信号是否能达到用传统金属电极采集的肌电信号 的控制性能仍然是个未知数。在本研究中，我们利用导电布制成的织物电极采集肌肉表面电信号，并且通过识别九个类型 的肘部和手部动作的情况来评估其性能。初步的实验结果表明：在四个正常人做实验对象的试验中，运用织物电极采集信 号的平均分类精度是94.34%，与用传统的金属电极的分类精度的94.25%没有明显的区别。我们的研究结果说明：织物电极 与传统的金属电极在多功能肌电假肢控制中的模式识别方面有相似的性能。 关键词 织物电极；多功能肌电假肢；肌肉表面电信号；模式识别 ABSTRACT　Surface electromyography (EMG) signals in myoelectric control of a motorized prosthesis are recorded commonly using the Ag/AgCl gel electrodes. Although a gelled electrode may provide high-quality EMG recordings, it is inconvenient in clinical application of a myoelectric prosthesis. A novel type of signal sensors, textile electrode, will be ideal for EMG signal recordings in control of a multifunction myoelectric prosthesis. However, it is unknown whether the performance of textile electrodes in control of myoelectric prostheses is comparable to commonly used electrodes. In this study, we used the commercial conductive woven to make textile electrodes for EMG signal recordings and investigated the performance of EMG signals in identifying nine arm and hand movements. Our results in four able-bodied subjects showed that the average classification accuracy across four subjects was 94.34% when using textile electrodes, which was almost same as that of 94.25% when using conventional electrodes. These pilot results suggest that the textile electrodes could achieve similar performance in classifying different arm movements for control of multifunction myoel