The project is currently in active development.
Space Station Tracker is a web application that helps you spot the International Space Station (ISS) from your location.
Did you know the International Space Station is visible to the naked eye? It appears as a bright, fast-moving star crossing the night sky. However, visibility depends on specific conditions:
- Time of day (typically at dawn or dusk)
- Your location
- Weather conditions
- The ISS's orbital position
The ISS is only visible when sunlight reflects off its surface while the observer is in relative darkness. These visibility windows typically last just a few minutes.
This application aims to:
- Calculate when the ISS will be visible from your specific location
- Send notifications before visibility events
- Provide additional information about the space station and its current mission
- Provide similar information about The Chinese Space Station (Tiangong)
- Display the trajectory as a path across a map of your local sky
The project consists of three main components:
- Node.js application written in TypeScript
- Handles orbital calculations and visibility predictions
- Manages TLE (Two-Line Element) data retrieval and processing
- Entry points:
src/index.ts: Main application for calculating ISS visibility timessrc/updateData.ts: Script for fetching and updating TLE data incore/data/norad-{satelliteId}.json
- Calculation logic stored in
core/src/ganit/directory
- Will serve as an API layer between the core calculation module and frontend
- Will handle location data, and notification preferences
- Will cache calculation results to improve performance
- Will cache the TLE's for ISS and other satellites
- React-based web application
- Interactive sky map showing the ISS's path
- User location management and notification settings
- Mobile-responsive design
- Demystifying the USSPACECOM Two-Line Element Set Format - Comprehensive explanation of TLE format
- Computers and Satellites - Primary resource for orbital calculation mathematics
- NASA's TLE API - Provides up-to-date TLE data for the ISS
- Additional NASA APIs - Other potential data sources for the project (only the above api is used in this project in the current state)
The project is in development, with the core calculation module as the primary focus. Current work includes:
- Creating a data refresh strategy based on TLE update frequency
- Figure out how often the TLE api updates
- Implementing orbital propagation algorithms (not the SGP4 but using SGP4 propagation of whether the space station is visible or not)
- Building test fixtures for validation against known ISS passes
- Node.js
- npm
# Clone the repository
git clone https://github.com/gopal-lohar/space-station-tracker
# Install dependencies for core module
cd space-station-tracker/core
npm install
# Run development version
npm run dev- TLE data retrieval mechanism
- Look Angle Calculation
- Sun Calculation and Testing
- Look Angle cacluation testing
- Is Illuminated calculation ( although it is done and works there is still some problem in it which is causing some wrong outputs)
- Is Illuminated calculation testing
- Complete visibility prediction algorithm
- visibility prediction testing
- Write Tests and actually verify the working against known ISS passes
- Backend API development
- Frontend application development
- User notification system
- Sky map visualization (MAYBE)
- some day the entire app can be converted into a react app with a web assembly worker thread
the current setup is somewhat upgradable using turborepo and can be extended to support more features.
but what is an interesting idea for the future? Maybe we can move all the computational logic to the frontend itself and do it using wasm and web workers. although not strictly required but it will be a good idea to use a better programming language for the computations, something that is fast, efficient, typesafe and better suited - Rust. Alongside changing programming language, the computations can themselves be optimized by plotting the graphs of the parameters and then using things like golden section search or maybe, just may be Newton's method as described in this article titled Sunrise . . . Sunset.
This project is licensed under the GNU General Public License v3.0 - see the LICENSE file for details.