Some satrecs return false for position/velocity with propagate about satellite-js HOT 8 CLOSED

Wanted to mention that the returning of [false, false] (after upgrading from 1.3.0 to 2.0.2) is an issue for myself as well

from satellite-js.

For example for decaying satellites sgp4 returns {position: false, velocity: false}.
satrec.error indicates if position is false.

In your case it seems to be satrec.error = 1; // sgp4fix to return if there is an error in eccentricity. So may you can just skip satellites with satrec.error > 0 or dive into SGP4 to figure out why. :)

Also note, that the TLEs are designed for a precise prediction within maybe two weeks. After that the satellite drifts to far away due to perturbations. Calculating the terms could fail in the future.

from satellite-js.

Ahh, my mistake. The TLE does load into satrec, but my application throws the error when converting to geodetic coordinates due to lack of position information.

satrec.error contains the integer 1 for problematic TLEs.

I looked at the data in a parallel coordinates view to see which orbital parameters were special for MICROMAS and FLOCK 1B-28. They both appear to have high Mean Motion and First Derivative of Mean Motion.

http://bl.ocks.org/syntagmatic/4446b31c6cd746eedaeb

Perhaps these are problematic for prediction? I'm wondering if there's some threshold to know when prediction by satellite.js isn't possible.

Thanks for the note on TLE prediction accuracy. I'll probably go back to the examples and have them start from the date the data was collected and loop after a week or two. It's mostly a simple example of how to use satellite.js with D3.js, I don't plan to keep the data up-to-date.

from satellite-js.

For your visualization you can may use weeks. But for example in that time it also can decay.

satrec.error contains the integer 1 for problematic TLEs.

satrec.error > 0. It's devided into different cases.

from satellite-js.

Those visualizations are really cool!

from satellite-js.

Yes, sorry. They are! :)

from satellite-js.

I don't think there's actually a bug in the calculations here, calculating near the epoch works fine (FLOCK 1B-28)

console.log(JSON.stringify(satellite.sgp4(satrec, 0).position));
{"x":-6486.244055534791,"y":1103.1672497210636,"z":-0.11404306548681002}

Eccentricity is a function of bstar and time, and when to try to propagate to current time, it's negative.

sgp4 should return { satrec: satrec, position: undefined, velocity: undefined } instead of [false, false] though.

from satellite-js.

I just wanted to toss out a workaround since regardless of returning undefined or false, you can't plot it. If you use a try/catch function (https://www.w3schools.com/js/js_errors.asp) then it will catch all of the false returns and you either skip that iteration or remove the satellite (99% of my failures are decayed objects that I need to remove from my catalogue).

from satellite-js.