Your new best friend.
Developing process!
Testing out the robot's swift reflexes!
Just a glimpse of the math involved.

AutoPong, your partner in ping-pong crime:

We know that sometimes, you just have that urge to play table tennis/ping pong.

We know that no one is available to battle it out with you.

We know that you're going to love the AutoPong.

What it does

The AutoPong is an automated robot that is able to return any moves that a player makes, able to keep up rallies alongside providing challenging shots for players to improve their skills with.

Inspiration

As table tennis casuals, we realised the extent to which we lacked in skill... In order to compensate for the lack thereof and to help improve the skills of our friends, we thought that such a tool would be invaluable!

How we built it

Using the OpenCV library, we utilised Processing to build a method by which a camera can detect the ball moving towards it. By means of converting the RGB values in the camera's feed into individual HSV channels and through the physics concepts of trajectory tracking and motion estimation of the ball, we were able to build a successful means of doing so.

Next, we used the arduino and hardware parts to build a robot skeleton and connect the two concepts, allowing them to work in harmony.

Challenges we ran into

It was really cold. Like honestly, it was surprisingly chilly.

Apart from the occasional craze that comes from caffeine and sugar overdose, we did face problems with the three axes when programming depth of object tracking as well as coordinate conversion from camera to real time. However, gradually we were able to get around these roadblocks.

Accomplishments that we're proud of

We are proud that the ball tracking actually works quite well! Adjusting the hue and saturation values took an unreasonable amount of time to get right.

The robot's infrastructure was also of a surprise to have been able to build within the short time that we had!

What we learned

Ambition has a limit. And time management is really important! We also learnt how to manipulate different, unfamiliar libraries within a limited amount of time.