jeb_ | Devpost

Inspiration

The recent Minecraft Movie and the craze for the hit game brought us together to recreate something from the iconic film that had functionality and presence, replicating the charming vibe, goofy aesthetic, and in game functionality of a Minecraft sheep.

What it does

This sheep does everything an in-game Minecraft sheep can do, including ambient "BAAAAA" noises, a love for wheat, following wheat until it reaches the target, and gives hearts to love you when it is fed.

How we built it

This sheep was built using 3 main embedded devices. The first is an arduino using a DFMiniPlayer and speaker to play between 3 in game audio files of the sheep's noises to represent the ambient noises you'd hear in the game. The next is the rover functionality of the sheep, which operates using an L293D motor driver connected to an arduino to drive the wheels of the rover (or the legs of the sheep) to move towards the wheat, turning and accelerating when necessary. This rover design is also assisted by an external laptop webcam using image recognition through a python script, which acts as a way for color detection. This color detection from laptop to sheep is ultimately how jeb_ knows to find the wheat. Finally, we use a tension pulley system connected to this arduino to pull up a love heart from the sheep's compartment, representing its love.

Challenges we ran into

The biggest challenges we ran into were the issues of connecting the sheep to wifi as we required a 2.4 gigahertz wifi connection. Without this vital connection, we could not get our sheep to transmit its target location and directions. The next biggest challenge was the issue of tension and torque force to get the heart to come up from the sheep properly. Finally, the greatest challenge was optimization of space to integrate each embedded system.

Accomplishments that we're proud of

The thing were most proud of is using color detection with an external camera and the makeshift pulley system used to raise this heart.

What we learned

What we learned most is how to utilize machine vision through color detection and version control through Git.

What's next for jeb_

Given more time, the project could definitely be more optimized and space allocated properly. Directionality could be better implemented in the future along with a better system to pathfind and recognize specific shapes and colors.