SpaceTracker

An AI-powered space situational awareness platform for tracking satellites, debris, and predicting orbital passes using real data from Space-Track and CelesTrak.

Features

• 3D Globe Visualization: See real-time positions of tracked satellites over a 3D Cesium globe with dynamic depth and altitude scaling.
• Agent-Driven Globe Interactions: AI responses automatically fly the camera to specific satellites and highlight queried satellite categories.
• AI Chat Assistant: Ask natural language questions about space, orbits, and celestial events using real-time API integrations.
• Location-Aware AI: Provides native web search and location context for celestial events without needing additional search API keys.
• Multi-Provider AI Fallback: Integrates multiple AI providers (Google, Anthropic, OpenAI, Groq, OpenRouter) and automatically rotates on rate limits or errors.
• Real-Time Data & Interpolation: Syncs with Space-Track API for the latest Two-Line Elements (TLE) and uses 60fps real-time ECEF velocity interpolation.
• Interactive UI: Glassmorphic UI with floating hover cards, detailed satellite info panels, and dynamic satellite imagery toggles.

Setup Instructions

Prerequisites

• Node.js v19+
• Docker & docker-compose
• Free account on Space-Track.org
• API key from an AI provider (Google Gemini, Anthropic, OpenAI, Groq, or OpenRouter)

1. Database Setup

Start the local PostgreSQL database using Docker:

docker-compose up -d

2. Install Dependencies

Install dependencies for both the server and the client with a single command from the root directory:

npm run install:all

3. Server Configuration

Create a .env file from the example in the server directory:

cd server
cp .env.example .env

Update your .env file with your Space-Track credentials, at least one AI API key (e.g., GOOGLE_AI_API_KEY), and optionally a Cesium Ion token:

SPACE_TRACK_USERNAME=...
SPACE_TRACK_PASSWORD=...
GOOGLE_AI_API_KEY=...
VITE_CESIUM_ION_TOKEN=... # Optional: Get a free token at cesium.com/ion for ultra-HD imagery

Run database migrations and seed the data:

npx prisma migrate dev
npm run seed
cd ..

4. Running the App

To start both the client and server concurrently, run this from the root directory:

npm run dev

On first load, the app will request your location to enable location-aware AI and pass predictions. If denied, a prompt will let you enter your city manually. Access the application at http://localhost:5173.