Inspiration
People tend to be bad drivers sometimes, even including the authors of this project. Therefore, we tried to come up with the way to increase self-awareness about their driving skills.
What it does
Our application tracks the user's drive manner and tells him how awful he is at driving. (literally, the larger index, the worse driver is).
How we built it
We used deep learning implemented with python (tensorflow) in order to create models of good driving and bad driving. The models were created from images taken with a camera attached to the front of the car. The idea of the app is to have used upload a video of their driving and the python script would generate the driver's score on an online. The app would have been hosted on node server.
Challenges we ran into
The most difficult part was to build a bridge between node.js and python, also we bumped into significant challenges with hardware. Our goal was to have the machine learning algorithm run along the side of the online app.
Accomplishments that we're proud of
We are proud to say that after our initial project of a remote control car that
What we learned
We learned that we should be careful with our hardware.
What's next for Loco Driver
Loco Driver will live on in our hearts.
Log in or sign up for Devpost to join the conversation.