Inspiration
The inspiration for this project is an easter egg from Portal 2! (spoilers)
Chell on bring your kids to work day, visits Aperture Science, where she and other children build Potato batteries, hoping that will become the future of energy. However, many, many years later, when the events of Portal 2 take place, as a little easter egg, you come across these projects, one of which was created by Chell, however, it has had an extremely odd growth, and has become a monstrous potato.
This project meets the theme, as the theme "Growth" fits with the growing of the potato. Also inspired by the nature of the resilience of Potatoes and the fact that the sprouts seem to violently grow, in search for soil.
What it does
It is a simulation of sprouts/roots trying to find moisture within soil. Plants, potatoes included, grow their roots by hydrotropism, meaning they involuntary grow towards moisture by detecting the water concentration within the soil. The simulation reproduces this affect in a completely randomised environment, as well as provides users a sandbox type environment where they can control some of the logic, parameters and scenario the simulation will run in.
How we built it
We built it using rapid prototyping, to see what algorithms worked and what didn't, and what resembled and was more accurate to growing roots.
Challenges we ran into
Finding the balance between realism and have it still be pleasant to watch was a task, but even more so was creating the randomised environment for the moisture blobs that spawn around, we decided to use perlin noise to randomly generate a pattern, which then is converted to a topographic map, however, instead of measuring altitude, it instead measures how moist the soil is, with the absolute peaks being 100% moisture (essentially pure water) and the absolute troughs being where the soil is completely dry with 0% moisture concentration.
Accomplishments that we're proud of
Creating something that is very satisfying to watch and play around with, It's quite interesting seeing the sprouts/roots violently searching for moisture.
What's next for Where is the SOIL?
Migrate the project to a incorporate a neural network derived from real data collected from growth patterns of Potato, but also other vegetables, and organic plants. Producing a model that will have a significantly higher accuracy to how the roots would actually grow. Allowing for mass data collection of potentially ideal breeds of plants and also ideal placing of plants when sowing the seeds.
Definitely implement it in a more efficient language, especially for training Neural Networks, like Python.
Please try it out with the link at the bottom, the site will open locally so you don't need to download anything from GitHub!!

Log in or sign up for Devpost to join the conversation.