GridCast

Have a safe trip, and let me know when you get there.

Comment

Our safe path up against the fastest path. What will you decide?

Inspiration

In Los Angeles, known for its traffic and crazy drivers with over 50,000 cars accidents a year, a safe path would be an great!

What it does

Visualize dangerous areas for driving in Los Angeles. Use AI to plot a safe route. Forecast the safest times of the day and week to drive.

How we built it

MapBox
Dash
Prophet
Flask
Boostrap
jQuery
Pandas

Challenges we ran into

Building a navigation algorithm that avoids danger zones, in particular finding a good heuristic and translating it to MapBox
Hosting our fbprophet machine learning model on gcloud was a challenge given that pystancould not bee installed.
Integrating Flask and Dash

Accomplishments that we're proud of

Successfully building the navigation algorithm

What we learned

Using the MapBox API
serving machine learning algorithms we trained ourselves

What's next for GridCast

Built With

Submitted to

LA Hacks 2020

Created by

Worked on data preprocessing, data analysis, and setting up the server environment for the application.

Oasys Okubo
Worked on the backend and set up routes for our application. Utilized Jinja2 to display interactive maps and pages. Helped design algorithms for safest path.

Garrett Leising
I worked on connecting our backend data with the mapbox map and adding the heatmap.
My biggest challenge was creating a secondary mapping system to allow for the generation of a route with safety as its priority instead of speed.
This was my first time working with mapbox and I really enjoyed the experience and look forward to creating more with it.

Jonathan Amar
Worked on data processing, forescasting accidents, building the accidents dashboard, and the flask backend to serve these components.

Jswig Poirel

Updates

Oasys Okubo started this project — Mar 29, 2020 09:47 AM EDT

Leave feedback in the comments!

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