Pneumatic Training Glove Based on Electromyographic Signal Control

In modern society, much public attention is paid to the rights and interests of the disabled, which become a social problem. In order to satisfy the pursuit of disabled for a better quality of life, it is necessary to give them a normal life. For example, people with degeneration or atrophy of hand muscles caused by accidents or other reasons need an auxiliary system to help them recover their hand muscles. However, after consulting a large amount of data, I found that most hospitals in the market use hard levers and springs as mechanical structures, which will result in excessive resistance, poor finger curls, and even aggravate the disease due to improper use. In contrast, the deformable structure is softer. Meanwhile, it can actively bend the fingers of patient through the inflation control method to achieve the purpose of safety and effectiveness. In this paper, the design and casting of pneumatic muscle are completed through independent modeling and pouring in the experiment [1]. At the same time, this paper also provides and tests two machine learning pneumatic muscle control algorithms based on real-time binning, which are idempotent and real-time, and can be fully controlled in combination with the designed pneumatic muscle. Its core signal comes from EMG signals on key muscle groups of the arm. The signal is collected from muscle in real time by wet electrode patch. For the control model of machine learning, we have tested and implemented a variety of gesture postures, including common palm curling and stretching, common finger movements, etc. The pneumatic muscle hand rehabilitation auxiliary glove based on electromyographic signal control has the characteristics such as passive exercise, safety, less pain, individuation, low price, full and effective exercise, etc. The pneumatic muscle-assisted rehabilitation system based on electrical signal control in this paper has broad application prospects and extremely high value.