Flight Tracking

[Mick West]

The de-facto TLE propagator for satellites in usual orbits is SGP4

I use the satellite.js library, which uses SGP4
https://github.com/shashwatak/satellite-js

satellite.js​

A library to make satellite propagation via TLEs possible in the web. Provides the functions necessary for SGP4/SDP4 calculations, as callable javascript. Also provides functions for coordinate transforms.

The internals of this library are nearly identical to Brandon Rhode's sgp4 python library.

Content from External Source

 

[FatPhil]

Old, as in more than a few days, TLEs are not very useful for reentry prediction purposes. See for example this comparison here,


(from https://analyticalgraphics.my.sales...s/HowTo/Can-TLEs-be-used-for-Decay-Estimation)

It's data for Iridium 96
set 1 - black/green - apogee/perigee from Jan 2020 to Jun 2020 derived from TLE valid Jan 1
set 2 - blue/purple - apogee/perigee from Jan 2020 to Jun 2020 derived from most current TLE
actual reentry occured on May 30, about where set 2 takes the rapid plunge

The extrapolation from just the Jan 1 TLE looks pretty good for a few weeks, maybe even a couple of months, not just "a few days", depending on your accuracy requirements. Of course, you should always know the limitations of the data and the model you use, but it doesn't seem as bad as you're making out.

 

[Easy Muffin]

The extrapolation from just the Jan 1 TLE looks pretty good for a few weeks, maybe even a couple of months, not just "a few days", depending on your accuracy requirements. Of course, you should always know the limitations of the data and the model you use, but it doesn't seem as bad as you're making out.

Ah, I could have been more specific - by for reentry prediction purposes I meant the comparatively short time span of the days preceding the eventual burn-up. In the graph it'd be shortly before the point at which the perigees begin to deviate.

 

[FatPhil]

Ah, I could have been more specific - by for reentry prediction purposes I meant the comparatively short time span of the days preceding the eventual burn-up. In the graph it'd be shortly before the point at which the perigees begin to deviate.

OK, understood, yup the exponential increase of air density towards earth does mean that the resistence ramps up as it approaches its final orbits, so any errors in your input result in larger errors in the output, and so your model gets significantly less predictive as the catastrophe gets closer.