Parkathon is a web application designed to help users find parking spots efficiently. Using user feedback and machine learning, it provides probabilistic predictions for parking availability, making it the perfect parking companion.
-
Browsers
- Some browsers like Chrome or Brave restrict geolocation because we have self-signed SSL certificate! (Alternatives like Microsoft Edge or Firefox work for geolocation).
- Some browsers like Firefox do not support Voice Recognition (Alternatives are Chromium based browsers).
TL;DR We recommend testing the application on Microsoft Edge.
- Real-time Availability
Determines parking availability based on nearby users' parking or unparking activities.
- Historical Data Analysis
Fetches historical parking data from a database and uses a machine learning model to predict future parking availability.
- User-friendly Interface
An intuitive frontend to display parking predictions to users through a map.
- Parking Status
Users can mark spots as parked or evacuated based on their vehicle’s location. Collected data is stored in a database and later used to train an AI model to predict future parking availability. Parking outside designated OpenStreetMap-based locations is considered illegal and is not recorded. Examples of illegal parking include spots near intersections, trash cans, or other restricted areas.
- Search for Available Spots
Users can search for empty parking spaces near a specified area using text or voice commands. The AI model analyzes unoccupied spots to estimate the likelihood of them being truly available.
How to deploy with docker-compose
docker-compose build
docker-compose up
For more information about the application you can view the docs/ directory.
This project is licensed under the MIT License - see the LICENSE file for details.