Hack Arizona 2020 submission, uses geolocation data of vehicles to provide safety updates, trend analysis, and live event presentation to the user. Data is used from several sources, and is available via a mobile app for increased safety.
Instructions for self-deployment:
Google Drive Link: https://drive.google.com/drive/folders/1to6_IomXK14IzFqfoJ_Bd7371cI9ROzV
Backend: Get the data from the NATS Read/interpret the NATS data Predict the path using the NATS data position/velocity Decide whether there will be collisions, close, near collision Send alert back to NATS (or wherever) Tunnel any Python data or outputs into the Node backend Alternative dataset that can be used for crashes
Frontend: Routinely pull live data from backend server (Ajax?) Update frontend display with live data Python script Execution based on routine or intialization Earth map overlay Static overlay of crash sites Popup modals or listing in page of scenarios (live) Popup modals or listing in page of suggestions (live) Dynamic/live view of vehicle position Dynamic/live line of prior path Dynamic/live line of predicted path Heatmat of crashes
Dependencies: Python 3.8 or later sudo apt-get install python3.8-dev pip install:
- asyncio-nats-client
- nkeys
- protobuf
- python3-protobuf
- geopy
- numpy