Simulation of Philae lander bouncing on the comet using Orbiter

[flug]

Hey all,

I've been using Orbiter to check out the best possibilities for where the Philae comet lander may have ended up after the Great Bounce it took the other day. You can check out the results here:

https://www.youtube.com/playlist?list=PLcoFChnR-MRhLMzzjO0cDNkx7AU1LBTM9

As a by-product, I made a quick mod of the nice Rosetta/Philae Add-on by Brian Jones, to allow the cool new Comet 67/P 3-D texture to be used in Orbiter, and also to update the config file slightly to represent the now-known rotation rate for the comet.

Result is a pretty cool experience flying to Comet 67/P.

It's available here: [ame="http://www.orbithangar.com/searchid.php?ID=6671"]Rosetta - Comet 67/P Landing Mission, ver. 3D[/ame]

Also via my own webpage here: http://brenthugh.com/orbiter/rosetta_v3D.zip

UPDATE: I released a version 3C of the Rosetta package already (link above). Main updates:

- Fixed the scale of Comet 67/P (it was a bit too small; now fixed based on accurate Rosetta mission data)
- Added/improved scenarios for Comet 67/P, fixed surface bases for 67/P
- Added ROT & LIN thrusters for Philae so that you can maneuver around the Comet, takeoff & land, control rotation
- Added lumpy gravitational map. Probably not very realistic yet but it definitely has some interesting quirks compared with a regular spherical gravitational well.

UPDATE 2: I just released V.3D of the add-on, with these interesting features:

- Corrected 67P spin axis and obliquity (now conform to latest science data)
- New scenario demonstrating VERY strange orbit of 67P created by fast spin + lumpy gravity.

Many thanks to BrianJ for creating the awesome Rosetta V.3 add-on. I've just added a few tweaks on top of that to incorporate ESA's most recent science data & mesh model. But his detailed and amazing work made it very easy to add those few new tweaks!

 

[BrianJ]

Ah! This looks wonderful! Will be downloading as soon as it's available on OH.
Thank you very much

 

[boogabooga]

If this is your first upload to O-H, you need to send a P-M to Vash. See this thread:
http://www.orbiter-forum.com/showthread.php?t=32445&highlight=Project+Gemini

 

[orb]

If this is your first upload to O-H, you need to send a P-M to Vash.

Vash is currently inactive at O-F or OHM, but Xyon has had OHM add-on approval powers for some time starting this year, too.

 

[boogabooga]

I saw the youtube video.

Are you using a custom gravity potential map?

 

[flug]

I saw the youtube video.

Are you using a custom gravity potential map?

No, it's just using a basic sphere for now. So that part is not completely realistic.

I thought just modeling the gravity as two spheres rigidly attached to each other--each one centered in each end of the comet, would probably be a major improvement. But I can't see that orbiter is capable of doing that. Does anyone know how to do that?

The other, perhaps more reasonable possibility for Orbiter would be to use a some parameters to generate a nonspherical gravitation field, as explained in the Orbiter docs at Doc\Technotes\gravity.pdf

As I understand that we can use that system to generate a gravitational field of an oblate object, maybe even something like a long skinny cigar shaped object. That might be something of an improvement over a simple sphere, but what we really need for Comet 67/P is more of a dumbbell-shaped object.

Anyone know if that is possible using the the Legendre polynomial series expansion approach used by Orbiter? My guess is, it isn't, but I don't understand the details enough to say for certain.

If anyone has any ideas about how to better model this, I'm all ears!

 

[flug]

Vash is currently inactive at O-F or OHM, but Xyon has had OHM add-on approval powers for some time starting this year, too.

Thanks for letting me know. I messaged Xyon but I also put the file on my download site below, so that you don't have to wait ;-)

http://brenthugh.com/orbiter/rosetta_v3C.zip

---------- Post added 11-20-14 at 09:56 AM ---------- Previous post was 11-19-14 at 03:47 PM ----------

The other, perhaps more reasonable possibility for Orbiter would be to use a some parameters to generate a nonspherical gravitation field, as explained in the Orbiter docs at Doc\Technotes\gravity.pdf

I've been thinking about this a little more. The 3C version includes some tweaks to the gravitational map to make it more two-lobed, with one lobe heavier than the other, etc.

However, there are a couple of problems with this approach:

- The fact that Orbiter requires you to designate a sphere radius and then treats that radius as the surface of the planet (ie, not allowing spacecraft to penetrate the surface) makes it difficult to use this approach to model an object like Comet 67/P. Ideally you would have a large sphere that encompasses the entire comet, then use the J factors in the gravity zone map to tweak the gravity field as needed. It could be like a two-lobed object with one lobe heavier than the other--a pretty good match. This approach won't assumes symmetry around the rotational access, so you could capture part of the asymmetry but not all of it. But MOST of it, I think.

But it wouldn't be much fun, or much use, to have a giant shell around the comet that no spacecraft could approach.

- Also Orbiter requires the sphere used in the zone approach to the gravitational field, to coincide with the rotational axis of the object. That doesn't work well for an object like 67/P. Ideally here you would use the long axis as the basis of your gravitational field. The axis of rotation is basically the short axis (so close to 90 degrees off from the long axis).

- Also I'm not sure how to generate the needed J factors, other than just guessing & experimenting.

Any thoughts, I'd welcome them! If there is some way to allow spacecraft to penetrate the surface sphere, that would be helpful.

 

[Xyon]

3B: [ame="http://orbithangar.com/searchid.php?ID=6670"]Rosetta - Comet 67/P Landing Mission, ver. 3B[/ame]
3C: [ame="http://orbithangar.com/searchid.php?ID=6671"]Rosetta - Comet 67/P Landing Mission, ver. 3C[/ame]

 

[flug]

I made another video showing the possible flight path of Philae:

https://www.youtube.com/watch?v=xXgcU5mvJhA&list=PLcoFChnR-MRhLMzzjO0cDNkx7AU1LBTM9&index=7

This is very close to the version that ESA thinks happened.

FYI I made some updates to the Rosetta/Philae add-on that improve realism quite a bit. I tracked down how to set the spin axis of the comet to the value as measured by ESA. That helps explain why some areas are as yet unmapped (they are in the southern hemisphere and it is 'winter' there since Rosetta has arrived--equinox is approaching and then summer, however . . . ).

Also, more important, I realized the axis of rotation was 90 degrees off. That is corrected and shown in the video above. It makes quite a difference, especially since the (correct) axis of of rotation is almost exactly in the same direction as Philae's final flight, if ESA is correct.

I'll have those and another few tweaks in a ver 3D release of the Rosetta add-on soon, I hope.

Also I have another and even more detailed mesh for Comet 67/P that will be interesting to try! It has detail on the back/dark area that is fairly blank in our current model.

---------- Post added at 02:34 PM ---------- Previous post was at 07:52 AM ----------

And check out this video of an ABSOLUTELY wacky, seemingly impossible orbit by Philae around the comet:

https://www.youtube.com/watch?v=ofwlIUld9s8

It happens by some combination of the fast spin of the comet (realistic) together with the WAG of the lumpy gravitational field I put together. The lumpy gravitational field certainly doesn't reflect the comet's gravitational field with fidelity, but it certainly does illustrate the type of thing that can happen under unusual/lumpy and fast-rotating gravitational fields . . .

Honestly when I stumbled across this yesterday, I couldn't believe what I was seeing.

 

[BrianJ]

Hi flug,
really appreciate the work you're putting into this. Love the crazy Philae orbit vid!

I was just updating a couple of the scenarios to go with the new 67P, but will hold off posting until you've done your thing with the rotation axis and mesh

While trying out the scenario "11 - Rosetta deploys Philae lander" I discovered that:

a) the jettison speed of Philae from Rosetta is too great - should be 0.19m/s
(the current jettison speed is greater than escape velocity!)

b) Rosetta must do some kind of "bombing-run" approach before Philae jettison, then correct it's orbit.
(from a circular orbit at 20km, you need an extra 0.5m/s dV directly towards the comet to get Philae down in the right amount of time)

In the meantime, here's an update for anyone who likes a more "gasseous" appearance to their comet. It replaces the static blue "coma" with an out-gassing module that generates a particle stream (see pic, .zip attached to this post)

Also, I have never experimented with making custom .dll's for a planet/comet but I might see if I can come up with some kind of double/triple planet(comet) to see if we can get a better match for the gravity field (and also just for fun!)

Cheers,
Brian

 

[flug]

FYI I just released V.3D of the Rosetta add-on, with these interesting features (which I had already mentioned upthread):

- Corrected 67P spin axis and obliquity (now conform to latest science data)
- New scenario demonstrating VERY strange orbit of 67P created by fast spin + lumpy gravity.

It's available here: [ame="http://orbithangar.com/searchid.php?ID=6671"]Rosetta - Comet 67/P Landing Mission, ver. 3D[/ame]

Also via my own webpage here: http://brenthugh.com/orbiter/rosetta_v3D.zip

Also FYI I'm leaving for Thanksgiving vacation later today, and we'll have no internet access for 9 days or so. So I'm done with updates for now. It's been very interesting and a great example of what Orbiter can do!

 

[BrianJ]

Attached .zip file has a couple of updated scenarios to go with Rosetta v3D

Scenarios:
10 - Rosetta orbiting 67P
11 - Rosetta deploys Philae lander

Also includes updated rosetta.ini with corrected Philae jettison speed 0.19m/s