Hey there Community! I’ve got another great project to share today, just released by YouTube creator James Bruton! For his latest project, James created an awesome automatic robotic arm prosthetic based on the DYNAMIXEL X-series.
The goal of this arm is to automatically assume poses based on the other positions of the wearer’s joints- in other words, the prosthetic uses its own “experience” to know what position to take. Using a machine learning setup, James trained the arm with the help of a body suit (the red and white “joints” in the video) to record responses:
Check out the full video and project below!
After training the arm and setting it up, it works as an impressive proof of concept! With data from three limbs, the arm works reasonably well to assume poses based on the wearer’s body.
With more fine-tuning, I’m sure it could be much faster and more accurate! Even so, this is an amazing demonstration of the potential innovations that some servos and a 3D printer can enable a determined maker to create!