A sEMG Brain-Machine Interface.

Description

I have been working with a friend on a BMI project since december 2019. We are true believers in the near revolution of brain machine interfaces (BMI) but BMIs available today are either invasive, constraining or awkward (in particular EEGs), and we believe surface EMG is the way to go. Although electromyography is an old technology, its potential for BMI has not been realized until recently and interest is growing. So we built an armband which possesses electrodes that are held in contact with the user’s skin. The electrodes, combined with a specialized acquisition/processing hardware, intercept and measure electrical signals propagating in the muscles when the user intends to move his/her hand (with or without actual movement). The processed data is transmitted via bluetooth to the device that the user wishes to control, and translated there into actual commands (moving a cursor, typing text, etc...). Please see our video !

Details

Our prototype is already able to generate state of the art recordings, but you guessed it, the bottle neck is how to make sense of all the data. We have tested several methods, from recurrent neural networks on the raw data to simple classifiers on processed data (sliding RMS of the signals, or sliding covariance matrix of the different channels) etc...

Do you have any idea on how to interpret sEMG signals into useful computer commands ? Contact us to get your hands on a prototype and implement your ideas.

https://www.youtube.com/watch?v=qNxASDsdeHg

Files

compil.mp4

This video demonstrates our current prototype's ability to control a simple video-game without having to move a finger.

MPEG-4 Video - 37.05 MB - 03/11/2020 at 19:33

Download

Project Logs

Collapse

Update
wolfgangouille • 11/02/2021 at 03:34 • 1 comment

Hello everyone !
First of all we would like to apologize for being away for quite some time now. Both Marius and I are quite busy with work, but we had a discussion a few days ago and we are resolved to release everything you need to build your own armband within a few weeks or months.
I believe we have been wasting some time trying to make the design more slick and compact, as there was still some plan to make a product out of it. However, now it is clear that what people want is something more DIY. We have been in contact with several research labs, which currently have very limited options for sEMG, as commercial system are either very expensive, or simply not available, and they would really benefit from a DIY sEMG armband. I'm sure the process of building the armband might be appealing in itself to many of you.
So we will make a few adjustments on the design so that it can be assembled by anyone with minimal engineering skills, without expensive equipment and over the span of a week end or so.
Files and instructions coming soon !
Renaud
New prototype
wolfgangouille • 06/13/2020 at 08:58 • 0 comments

After several iterations we arrived at this fully integrated design, based on the tiny FPGA BX.

View all 2 project logs

Discussions

perlinwarp wrote 09/22/2021 at 14:34

I've started a discord for people working on EMG projects here: https://discord.com/invite/mG58PVyk83

There's a few people making custom hardware here and talking about things such as what instrumentation amp to use etc

I've been working on open source EMG projects for the last year including a cross platform multithreaded driver for the Myo, the wiki contains general info about EMG too: https://github.com/PerlinWarp/pyomyo/wiki/The-basics-of-EMG-design

Are you sure? yes | no

hut wrote 08/16/2021 at 12:35

Really interesting project :). I'm building something similar [https://psylink.me], and would love to exchange perspectives and experiences with you! Drop me an email at hut ໑ hut : pm

Are you sure? yes | no

erp3300 wrote 04/16/2021 at 22:36

I'd love to buy a prototype! Do you have a venmo/paypal?

If not, can you send me your design, and I can assemble it

Are you sure? yes | no

perlinwarp wrote 02/04/2021 at 09:57

Made an account just to say this is really cool!

I'm a data scientist but would love to make one, can we get the details?

Are you sure? yes | no

Sasja Vandendriessche wrote 12/31/2020 at 10:01

How's the project coming along?

Are you sure? yes | no

Gitnock wrote 03/22/2020 at 10:12

how much would the finished build cost?

Are you sure? yes | no

wolfgangouille wrote 03/22/2020 at 22:08

Well, depends what the finished product is ! Right now a full prototype costs about $500 in materials. We want this to remain accessible to hobbyists so we are working on making this cheaper.

Are you sure? yes | no

Gitnock wrote 03/22/2020 at 22:38

Thanks so much for the reply. ＼(◎o◎)／ am super excited for a time I could get my hands on this. good luck

Are you sure? yes | no

Drew Pilcher wrote 03/21/2020 at 08:50

Extremely cool, but you should share your design.

Are you sure? yes | no

wolfgangouille wrote 03/21/2020 at 18:29

Yes, we are making it a bit more easy to assemble for the layman first :)

Are you sure? yes | no

A sEMG Brain-Machine Interface.

Description

Details

Files

compil.mp4

Project Logs

Collapse

Update

New prototype

Discussions

Similar Projects

Diabetes Data, Everywhere

Exoskeleton Simulator

ÓSK Squirrel

The Human Connection : 1st Impression

A sEMG Brain-Machine Interface.

Become a Hackaday.io member

Just one more thing

Description

Details

Files

compil.mp4

Project Logs Collapse

Update

New prototype

Enjoy this project?

Discussions

Become a Hackaday.io Member

Similar Projects

Diabetes Data, Everywhere

Exoskeleton Simulator

ÓSK Squirrel

The Human Connection : 1st Impression

Does this project spark your interest?

Report project as inappropriate

Send message

Remove Member

Project Logs

Collapse