Auto-Swish Money Hoop

Inspiration

This project was inspired by the "Automatic Bullseye, Moving Dartboard" project courtesy of former NASA mechanical engineer and popular Youtuber, Mark Rober. Also a tribute to the late, great. 5-time NBA champion and legend, Kobe Bean Bryant.

What it does

The premise of the project is to detect then predict the movement of a projectile using computer vision and kinematics.

How I built it

This was built primarily using arduino and opencv programs. An Arduino Uno for the motor control while the camera data was processed directly on a computer due to a lack of raspberry pis being available to us.

Challenges I ran into

One of the major issues we ran into was integrating the arduino, python and opencv programs and ensuring they run efficiently so that the net has enough time to react to the ball.

Accomplishments that I'm proud of

We are really proud of the learning experience and being able to work on a fun project like this, with equally challenging software and hardware obstacles. Optimizing the system was really important and we're really proud of the work we did to achieve this.

What I learned

Throughout, we learned a lot about machine learning, data/image processing and computer vision.

What's next for Auto-Swish Money Hoop

The next step would be to increase the degrees of motion of the hoop to 2 axes. The use of 2 cameras would allow us to track the coordinates of the ball in three-axes, which would allow us to throw the ball into the net and have it self-correct accurately.