[fractallyte]

"Our cameras in Northford, Connecticut, and Douglassville, Pennsylvania, as well as a doorbell camera in Wayne, New Jersey, captured the meteor, and from that we measured its trajectory... The path traced back to low in the asteroid belt."

I don't doubt it, but that's an astounding trajectory estimation...

[defrost]

Desert Fireball Network, and less public precursor projects have been doing this for a while now:

* https://dfn.gfo.rocks/

* https://en.wikipedia.org/wiki/Desert_Fireball_Network

* https://pawsey.org.au/researchers/desert-fireball-network-te...

[imoverclocked]

From the wikipedia link:

> The dark flight trajectory of a meteoroid is significantly affected by the atmospheric winds, especially by the jet stream. As a result, the meteorite fall position can be shifted by up to several kilometers compared to a scenario with no winds.

That network is mostly interested in where it falls, not where it came from. A few degrees for finding an origin is a huge variation.

[defrost]

Planetary scientists are motivated to physically examine freshly fallen near pristine samples, yes. They are also keenly interested in origins, intra or extra solar system, relationships to known showers et al.

The less public precursors were and still have an interest in both projected targets and sources.

[anovikov]

This is absolutely trivial as long as any fixed ground objects are visible in a camera and does not at all require some sub-pixel precision.

[onion2k]

It's trivial to draw a line between some fixed points and extrapolate back to an origin if there's no external forces acting on the asteroid. But there is - planetary gravity, the moon, potentially the sun heating one side of it, atmospheric wind, weather, etc make a big difference. That makes it harder.

[anovikov]

That's basics of meteor observation. As long as camera didn't move, you can plot the track against the stars and time is known. Then with two tracks from different locations, orbit is determined.

[lazide]

Also, a typical consumer camera is going to struggle (even with the best analysis!) to have visual accuracy within even a few degrees of true (due to everything from high compression ratios in post processing to shitty optical properties in the actual camera) and that makes a probability cone that is pretty large.

Still better than nothing, though!

[anovikov]

building a WCS model (pixel XY <-> ray vectors) of a camera is a solved science. Of course you can never expect it to be rectangular.

[lazide]

No kidding, that is what I was referring too. Precision for a typical doorbell camera is going to be several minutes of angle at best, which is my point.

[anovikov]

Oh well several arcminutes is excellent. Yes this is best case.