Open-source Python tool for checking NEO anomalies (aNEOS)

Showcase
1

Topic Title:
Open-source Python tool for checking NEO anomalies (aNEOS)

Hi all,

I have been playing with public NEO data for a bit and built this small Python project called aNEOS.

It fetches close-approach records and orbital elements from Horizons, SBDB, MPC and similar sources, then looks for objects that behave a little strangely. Things like unusually regular repeat approaches, low relative velocities that keep recurring, non-gravitational acceleration without clear outgassing, or radar/thermal/spectral hints that do not quite match typical asteroid groups.

So far everything gets classified as natural after the full checks, but the pipeline is sensitive enough to pick up the known edge cases quite reliably (1998 KY26 shows up in longer runs, for example).

What works at the moment:

  • Rich CLI with progress bars, detailed clue views and candidate browsing
  • 200-year historical polling of NASA close approaches
  • Basic impact probability and keyhole estimates
  • REST API for scripting if you prefer that
  • Caching to avoid hitting the APIs too hard

Repo: GitHub - RobLe3/aneos-suite: Open-source statistical framework for detecting artificial Near Earth Objects and computing planetary defense impact probabilities. 14-option CLI, REST API, 200-year historical pipeline, BC11 population analysis. · GitHub

It is just a research experiment, not built for alerts or official use. Always verify anything it flags against JPL, ESA or MPC data.

I wonder if anyone else has come across similar odd objects in their own data and thought “that looks unusual”. If you have, or if you enjoy tinkering with orbital anomaly detectors in Python, feel free to glance at the code and point out what is broken, too noisy, too conservative, or worth adding (for example hyperbolic orbits or interstellar objects).

Thanks for any thoughts or second opinions.

zKelvinM aka Rob