Van Volkinburg Kyle, Washington Gregory
IEEE Trans Biomed Circuits Syst. 2017 Aug;11(4):900-909. doi: 10.1109/TBCAS.2017.2683458. Epub 2017 May 29.
This paper reports on a wearable gesture-based controller fabricated using the sensing capabilities of the flexible thin-film piezoelectric polymer polyvinylidene fluoride (PVDF) which is shown to repeatedly and accurately discern, in real time, between right and left hand gestures. The PVDF is affixed to a compression sleeve worn on the forearm to create a wearable device that is flexible, adaptable, and highly shape conforming. Forearm muscle movements, which drive hand motions, are detected by the PVDF which outputs its voltage signal to a developed microcontroller-based board and processed by an artificial neural network that was trained to recognize the generated voltage profile of right and left hand gestures. The PVDF has been spatially shaded (etched) in such a way as to increase sensitivity to expected deformations caused by the specific muscles employed in making the targeted right and left gestures. The device proves to be exceptionally accurate both when positioned as intended and when rotated and translated on the forearm.
本文报道了一种基于可穿戴手势的控制器,该控制器利用柔性薄膜压电聚合物聚偏二氟乙烯(PVDF)的传感能力制造而成,能够实时反复且准确地辨别右手和左手的手势。PVDF粘贴在前臂佩戴的压缩套筒上,形成了一种灵活、适应性强且高度贴合身体形状的可穿戴设备。驱动手部动作的前臂肌肉运动由PVDF检测,PVDF将其电压信号输出到一个基于微控制器开发的电路板,并由一个经过训练以识别右手和左手手势所产生电压曲线的人工神经网络进行处理。PVDF已在空间上进行了阴影处理(蚀刻),以提高对由做出目标右手和左手手势所使用的特定肌肉引起的预期变形的敏感度。该设备在按预期放置时以及在前臂上旋转和平移时都表现出极高的准确性。