PLEASE SEE FULL VIDEO ANIMATION AT YOUTUBE!

https://www.youtube.com/watch?v=3MI8FgxQxsI

This image and animation were shot over the course of four nights, February 4th - 7th, 2021. I had only intended to capture the Rosette Nebula and lucked out into getting four consecutive nights of roughly similar amounts of time of the Main Belt Asteroid (130) Elektra skimming past this famous nebula. The time-stamp is UTC.

Please see the first revision here for a still image with the asteroid as a trail.

In order to integrate and incorporate the asteroid, I used the following steps. (Many intermediate steps are not included, this list is a basic reference...)

1) WBPP all subs, using LN and rejection, separate RGB, recombine. Process as usual.
2) Subtract unprocessed RGB recombined integration from all subs using Image Container and Pixel Math. This leaves all ephemeral stuff, including the asteroid. (Planes, satellites, cosmic rays, etc...)
3) Integrate "ephemeral stuff" subs using "Maximum" (rather than "Average"). (Thanks to @Adam Block for this, and the previous step, advice!) 
4) Create mask to protect only asteroid trail. Clip the rest of the image to zeros using HT.
5) Minor processing of remaining asteroid trail, add to processed Nebula integration to get final still image, first revision of this post: https://www.astrobin.com/8a6qh5/B/ (addition, which puts the asteroid in front of any brighter stars unlike "max")
6) Use Image Container with the mask of the asteroid trail mentioned above (to protect the asteroid) to clip all "ephemeral stuff" subs with HT.
7) Use Image Container to add each sub with only asteroid remaining to the processed nebula image using Pixel Math, creating 399 frames for prep animation.
8 ) Use Image Container and Process Container with Image Solver and Annotate Image to add platesolved information and date/time stamp.
9) Various futzing in Photoshop to crop and animate. PLEASE SEE THE YOUTUBE VIDEO!!! https://www.youtube.com/watch?v=3MI8FgxQxsI

As I was doing this my head was spinning with different ways I could have proceeded to be more "pure". 

I could have separated each of the "ephemeral stuff" subs into RGB, and realigned them to the same sub as the main nebula image to address some CA issues, (I'd have to think hard about specifically how to do this), but I chose to desaturate the asteroid a little to deal with that. It's basically grey-white, so the result is acceptable I think.

The asteroid is actually Apparent Magnitude +12.1 at the time this was shot. I have represented it arbitrarily a bit too bright I think, but that's inline with what we all do with star reduction and nebula stretching. I could have recombined the asteroid trail with the nebula image pre-stretching, but when I removed the stars and stretched the nebula, the asteroid's brightness would have been arbitrarily increased, (probably more than I ultimately did), relative to the stars anyway. I could have extracted the stars unstretched, and added the unstretched asteroid to the stars and adjusted it there, but I think it would have been made too dim when I reduced the stars. The asteroid's brightness relative to itself, as affected by prevailing atmospheric conditions and such, is correct. There are a lot of things to consider with this and other ways I might have proceeded, but I think I have successfully told the story of the time I accidentally captured a major asteroid passing a famous nebula in a compelling and convincing way. 

This was my first foray into using image and process containers to do image processing on numerous subs/images. This is such an amazingly useful, and easy to use, feature of Pixinsight! Adam Block Studios video tutorials were key to understanding the basic operation of these tools. Go sign up for the Fundamentals courses!