SatMon
Initializing Satellite Monitoring System...
Loading components...
Initializing Satellite Monitoring System...
Production-ready ML system for detecting spacecraft anomalies using NASA-style telemetry data. Built with FastAPI, PostgreSQL, and advanced machine learning algorithms.
Z-Score statistical analysis, Isolation Forest ML algorithms, and LSTM autoencoders with comprehensive performance evaluation against ground truth data.
Interactive web dashboard with live telemetry charts, anomaly overlays, and performance metrics for comprehensive spacecraft monitoring.
FastAPI backend, PostgreSQL time-series database, Docker containerization, REST API endpoints, and automated deployment pipeline.
Synthetic telemetry data that mimics real NASA JPL spacecraft patterns with realistic anomaly injection and ground truth labeling.
Comprehensive evaluation framework with precision, recall, F1-scores, and window-based overlap detection for rigorous performance analysis.
Fully containerized with Docker Compose, environment configuration, health checks, and automated data loading for production deployment.
International Space StationData sources: NASA ISS API, SpaceX API, NASA Mars Photos API
Real-time monitoring of system performance, database health, and API metrics for production operations.
Comprehensive API documentation with interactive examples and real-time testing capabilities.
Retrieves real-time data from NASA ISS API, SpaceX API, and Mars Perseverance rover.
{
"status": "success",
"iss": {
"latitude": 45.23,
"longitude": -122.45,
"altitude_km": 408,
"crew_count": 7
},
"spacex": {
"mission_name": "Starlink 6-1",
"flight_number": 251,
"success": true
}
}Returns live machine learning model performance, feature importance, and anomaly detection results.
{
"model_performance": {
"isolation_forest": {
"accuracy": 94.2,
"precision": 92.1,
"f1_score": 0.918
}
},
"feature_importance": {
"temperature_variance": 0.342,
"pressure_rate_change": 0.289
}
}Executes anomaly detection algorithms on specified telemetry channel.
{
"channel": "demo_channel_1",
"algorithm": "isolation_forest"
}Live analysis of actual satellite telemetry data using production ML algorithms. Real Isolation Forest and Z-Score detection running on demo_temp.csv with genuine performance metrics.
Machine learning model interpretation showing which telemetry features contribute most to anomaly detection decisions.
Real-time 3D visualization of the International Space Station orbiting Earth. Interactive globe showing current ISS position, orbital path, and ground track.
ISS Status
Interactive 3D Earth β’ Live ISS Tracking β’ Mouse to rotate and zoom
Select a spacecraft channel and run real-time anomaly detection. Watch as our algorithms identify anomalous patterns in satellite telemetry data.
Computer Science student at University of Colorado Boulder with a focus on Machine Learning and Systems Development.
Experienced in full-stack development, data science, and systems level development through classes, internships and personal projects.
Proficient in Python, C, JavaScript, SQL with experience in FastAPI, React, PostgreSQL, Docker, and machine learning frameworks.
Passionate about building production-ready systems that solve real-world problems.
