Every now and then you discover a story that just proves the power of making. This is the one that I will remember most for 2016, and I discovered it on Hackaday. It begins with a spunky girl, Lorelei, that suddenly gets a rare illness called AFM (Acute Flaccid Myelitis) which is similar to polio. She is lucky enough to be alive, but she is left with an almost paralyzed arm. While her father Bodo Hoenen agonizes over the next move, he realizes that hope for some sort of assistive device isn’t good enough. They would need a solution that is affordable and within reach, which means together they would need to DIY a robotic arm. Oh yeah, and they have “little to no robotic experience.”
Hackaday describes the process, which was presented in detail by Bodo at their Hackaday Super Conference.
They received help in learning how to 3D scan Lorelei’s arm for the best fit. A company in Canada sent them some actuators that met their weight, size, and power constraints. When they tested out the EMG sensors they discovered the signals on her damaged arm were almost completely lost in the noise. Again, help came in from a company working on a very similar problem. A new seventeen-sensor method was adopted that uses machine learning to find those signals. Most recently they have turned this into a video game — Lorelei is retraining her body by moving an arm onscreen.
I’m loving the flat 3D print for the wearable base, but my fave moment is the discovery made by Bodo and his daughter concerning picking up weak signals from her arm. Here’s the detail from Bodo’s blog.
I had been explaining to Lorelei that her muscle signals were like trains traveling down a track and that we could not get a reliable signal because the track was broken… A few days later, while at a train station, she looked at me with a spark of genius and said:
“Why are we only looking for a train”
That’s it! Let’s not filter out and normalize the signal, let’s look at the full signal.
This insight was what led them to use machine learning for the sensors. It’s really great to see a father and daughter working so well together and I’m sure Lorelei will never forget that smart moment. Here’s their update on the arm in action.
As you can see, this duo is facing their challenge using an open source collaborative approach. In fact, they are inviting the world to get involved as they are still looking to get help with machine learning, so check out their Facebook @OurKidsCanDoAnything. I’m sending huge hugs to Lorelei and Bodo for not only working on their own solution, but for empowering people around the world to realize the strength in open source work. Keep the spirit moving. For those that want to learn more about muscles and wearables, you should check out our guide on Getting Started With Myoware Muscle Sensor. Find out how muscles can control electronics.
Every Wednesday is Wearable Wednesday here at Adafruit! We’re bringing you the blinkiest, most fashionable, innovative, and useful wearables from around the web and in our own original projects featuring our wearable Arduino-compatible platform, FLORA. Be sure to post up your wearables projects in the forums or send us a link and you might be featured here on Wearable Wednesday!