Dan Nguyen & Rachel Platt
We wanted to make an accessible vehicle that would be controlled by a myoband. The controls for the vehicle are all only controlled by the motions of one hand/ arm.
For our input, we used the Myoband. We used the myoband detection of the hand and arm at different postions and movements for our training. We trained these to correspond to different vehicle motions: going at a constant pace, speeding up, braking, and jumping (because our car can fly). We chose to use K nearest neighbor for our project. When training, we had some difficulty with the Myoband recognizing changes in guesture, but this one seemed to work the best.
Make this on a 3d plane! We struggled to do this as there are so many extra things to consider on a 3d plane and constructing a successful output program, but it might help our project become a lot more realistic as an accessiblity vehicle.