Title:

Gesture Recognition by Flexible sensor array

The goal of this project is using EMG sensor arrays and one microcontroller to collect data, after the data collection, use computer to do training. After getting the trained data, the computer can understand the input you give, and predict what the gesture is. For this project, the most important and difficult step is collecting data and transfer data. You have to use some device to deal with the front end of the signal. Then the microcontroller and computer can understand the data from muscle electrical signal. In this video, there are 3 devices will be introduced to used to collect data. In the beginning, ADS1298 is a good choice. It has multiple channels and high accuracy, but the code used to connect with microcontroller to transfer data is very hard. The second choice is ADS1299, it has public library, the code is much easier. However, the data collected is not stable, this device is too sensitive, it will give you two different graphs when you di same gesture. In the end, the MyoWare Muscle sensor is a good choice. But this deice only have one channel. The accuracy is not very high. But It’s much easier to use and the code is much easier to write. As a result, the trained model is completed by Matlab. Now, if you give one input, the gesture can understand what it is.

Name

Comment*

Comments are viewable only by submitter

Captcha*