Inspiration

Astrowaves was born based on the not-anymore-so-far-fetched idea that space components and actuators, such as rovers, rocket (and rocket launches), and space suit functions can be controlled using brainwaves to increase accuracy, flexibility, and success rate of space exploration ventures.

What it does

It does exactly what you think it does :) Astrowaves made it possible to control physical space ships, rovers, rockets, and space suits using someone’s brainwaves.

How we built it

Astrowaves collects brainwave data using an electroencephalogram (EEG) headset. This stream of data is transmitted via Bluetooth to a flask server written in Python that uses the Google Cloud AutoML API to train and deploy a machine learning model based on the operator’s current stream of brainwaves. Furthermore, we strived relentlessly to create a machine learning model that would accurately and beautifully associate brainwaves patterns to actual physical commands in order to control the rovers, rockets, and space suits. Additionally, the entire front-end of Astrowaves was built from scratch using nothing other than the powerful Fantastic Four: HTML, CSS, Javascript, and Adobe Illustrator.

Challenges we ran into

Oh boy. The first challenge we came across was the fact that we did not have enough money to buy a full-fledged space suit ($12,000,000) a rocket, or not even a rover. Therefore, we decided to create a simulation on the browser that mimics real-life space exploration situations. Another challenge we ran into was creating a machine learning model that, over time, associates brainwave patterns to physical robotic commands. Finally, we also would have loved to have introduced 3D models to Astrowaves where rovers and rockets could choose different planets to go on the simulation, but we didn’t have the time for that because we decided to focus on the model’s accuracy and reliability.

Accomplishments that we're proud of

The Astrowaves team is very proud to create such an evocative platform. From the front-end to the dangerous wildernesses of the back-end and our datasets, we carefully designed a trustworthy and plausible platform that allows you to simulate real-life space exploration situations and control them with your mind. It’s like silk against your skin.

What we learned

The Astrowaves team learned a lot in such a small amount of time. The team had never tried to train a machine learning model that associates brainwaves patterns to control rovers in the moon or a CO2 dispenser in a spacesuit. That sounds insane! But we did it :)

What's next for Astrowaves

_ More. Waves. More. Space. More. Everything. _ The Astrowaves team would like to add 3D models of planets and moons to Astrowaves, where operators can be put into different circumstances and understand how their brains will differently control rovers, rockets, and spacesuits in different environments. We would also love to improve Astrowaves’ accuracy on the brainwave machine learning classification.

Github Repo:

https://github.com/maufcost/MakeUCSpaceExpo

Built With

Share this project:

Updates