Polaris: Open Source Machine Learning for Spacecraft Operations
summary
Polaris is an open source Python pipeline that fetches SatNOGS telemetry, augments it with orbital and space weather data via vinvelivaanilai, trains hundreds of XGBoost models to infer weighted dependencies, and serves an interactive 3D dependency graph in a browser. It processes LightSail 2 telemetry as a demo, has ~600 commits from 11 contributors across five countries, provides CLI commands fetch/learn/viz, and proves useful for spotting correlated signals and flagging anomalies.
AI-written summary · about
chapters (6)
related talks
24mRivian: Scaling Data & ML Platforms
31mPolars, the End of Spark?
33mPython VS Space Junk
21mPolars: A highly optimized dataframe library
06mMETASAT Demonstrator: Mixed Criticality, Accelerated AI Computing for Future Space Systems
45mFrom Radar Echoes To Real-Time Recognition: Deep Learning on Another Planet an Open Source Workflow
28mHow I used Polars to build built functime, a next gen ML forecasting library
31mFrom Radar Echoes To Real-Time Recognition: Deep Learning on Another Planet -An Open Source Workflow