Board index FlightGear Development Spaceflight

Space Shuttle - Development

Discussion about development and usage of spacecraft

Re: Space Shuttle - Development

Postby Thorsten » Fri Feb 26, 2021 6:28 am

But, I miserably fail to get Eileen doing stationkeeping, and the reason seems to be that the HST position and relative velocities set are the one of the orbital HST target and not the ones I set in the proximity zone using in the xml scenario file


I'm not sure I understand correctly, but: Is the issue that you want to instance an HST object in the proximity zone via the scenario code while there is an existing orbital target with the same name in the far zone that never got de-registered because you never entered the proximity zone?

That indeed won't work - the scenario is a quick way to enter action, it doesn't connect smoothly to the full mission concept. So If you want HST as scenario, delete it from the mission file and I expect it'll be okay.

(Come to think of it, Eileen just might reference the correct object if you tell her once more to change the reference object? But I'm not sure...)
Thorsten
 
Posts: 11991
Joined: Mon Nov 02, 2009 8:33 am

Re: Space Shuttle - Development

Postby eatdirt » Fri Feb 26, 2021 8:58 am

So If you want HST as scenario, delete it from the mission file and I expect it'll be okay.


I was really looking for the simplest thing, exactly as for ISS-docking or Spartan capture, I don't have HST set in the mission file.

For unknown reasons, once the 3D model appears at 100 ft from the Shuttle, Eileen finds it in the list of nearby object, but she's targeting another HST object (not present before switching on Eileen, and not set in the mission file). I can see that in the RDV mission, an HST appears, but it is thousand miles away. It is as if Eileen initiated the "orbital target HST". I would expect the situation to be as for ISS (which works fine). But maybe some funny effects appear because we can have HST as a payload and an orbital target?
eatdirt
 
Posts: 829
Joined: Wed Aug 15, 2018 2:06 pm

Re: Space Shuttle - Development

Postby Thorsten » Fri Feb 26, 2021 1:31 pm

But maybe some funny effects appear because we can have HST as a payload and an orbital target?


Meh - I've only ever tested HST as scenario, so that used to work okay, but it wasn't grabbable then, so I never tried something more fancy. I've also never deployed it, that was Wayne's baby.

So I can't rule out a funny effect - but people have been flying to HST?! So something used to work... I'll apparently have to look into this, but I'm still flying de-orbit and entry tests at the moment.
Thorsten
 
Posts: 11991
Joined: Mon Nov 02, 2009 8:33 am

Re: Space Shuttle - Development

Postby Thorsten » Fri Feb 26, 2021 1:48 pm

I've now pushed the notepad - the first 8 lines are writable by the scenario file, the lines 10-16 are used to log whatever MCC answers, it's filled bottom-up, so it should always be in the right order even when the answer is multi-lined.

It seems GinGin recorded PEG-4 insertion parameters inside some of the mission files - I guess what we should do is make the mission file be able to write the notepad as well so we can have the relevant parameters in a natural way in-sim.

Anyone who wants to see a demo - the de-orbit scenario allows some time to make the Shuttle ready for the OPS 3 transition (if you're not too slow), then burn the parameters as written on the notepad (actually a bit more OMS propellant could be burned...) and land in KSC.
Thorsten
 
Posts: 11991
Joined: Mon Nov 02, 2009 8:33 am

Re: Space Shuttle - Development

Postby GinGin » Fri Feb 26, 2021 2:17 pm

I am looking forward to test that next week.
Could you post a picture of it ?

For the peg 4 , I don’t think they are relevants for mission file ( some older values I think )
That is a good idea, at least to provide oms 2 peg 4 for ISS and hst mission files.
I have the exact parameters somewhere ( they are almost the same than real ones )
Tig at apogee ( around 38 mn) , c1c2 nul, and theta T around 315 degrees if I remember well .

For ISS

Image
GinGin
 
Posts: 1359
Joined: Wed Jul 05, 2017 10:41 am
Location: Paris
Callsign: Gingin

Re: Space Shuttle - Development

Postby Thorsten » Fri Feb 26, 2021 5:50 pm

That is a good idea, at least to provide oms 2 peg 4 for ISS and hst mission files.


Do they actually compute in-sim? (The PEG-4 solver isn't quite as good as the one of LEO targeting..., it doesn't always converge :( Yet another to-debug item...)
Thorsten
 
Posts: 11991
Joined: Mon Nov 02, 2009 8:33 am

Re: Space Shuttle - Development

Postby GinGin » Fri Feb 26, 2021 8:23 pm

I increased a bit the number of iterations for in sim peg4.
I will check the output with the standard number of iterations . I remember that the peg 4 values above where converging quite fast / around 10 iterations for ISS mission file and direct insertion parameters associated .
GinGin
 
Posts: 1359
Joined: Wed Jul 05, 2017 10:41 am
Location: Paris
Callsign: Gingin

Re: Space Shuttle - Development

Postby eatdirt » Fri Feb 26, 2021 8:26 pm

but people have been flying to HST?! So something used to work


Yep, I did a few times (before we changed hst.xml). But it was not grabable, so this has never been tested. I also tested as a payload and pull it out of the bay, but never a retrieve and all that is not the same chain of events as a servicing mission.

I have another beginner question, hst_manager() is defined both in rel_orbital.nas and in payload.nas; how these guys behave /overload when called as SpaceShuttle.hst_manager.amethod()? (sorry if that is trivial...)

Edit: Holy... I am getting it... I've just fixed all the problems by renaming the 2 guys in hst_payload_manager() and hst_tgt_manager(). These guys have nothing to do one with the other. So, for sure, the HST target is not grabbable.. :(
And that's quite some work to make that working... Either we would need an handover of orbital target to payload object close to the Shuttle, or may be, a merging, or inheritance, of the payload class from rel_orbital...
All payloads could indeed be orbital targets, but not all orbital targets could be payloads. WDYT?
eatdirt
 
Posts: 829
Joined: Wed Aug 15, 2018 2:06 pm

Re: Space Shuttle - Development

Postby Thorsten » Sat Feb 27, 2021 7:05 am

I didn't implement 'HST the payload', so I can say very little about the specifics, but

And that's quite some work to make that working... Either we would need an handover of orbital target to payload object close to the Shuttle, or may be, a merging, or inheritance, of the payload class from rel_orbital...
All payloads could indeed be orbital targets, but not all orbital targets could be payloads. WDYT?


Which is why I've been coding that for two months, so it should all be there. You can fly the SPARTAN into space and release it, you can grab it again, you can launch to its orbit, grab and stow it and land with it,...

We have all these handovers in place, it just seems they're not called the right way.

There's four different situations that need to be addressed:

1) Object secured in the payload bay - that's a child model of the Shuttle that's simply usually not shown and a mass point definition at the right place.
2) Object attached to the RMS arm - that's a second child model coming with the relevant set of translations/rotations to follow the arm
3) Object in the near zone - that's a complicated Nasal class handling motion, attitude, contact, tracking information and interaction with the Shuttle (docking, grabbing)
4) Object in the far zone - that's a simple Nasal class providing tracking position and tracking info.

The 1-2 handover is done by the RMS arm code, the 1-3 handover would be accidential, but is done by the latching code (if you release the latches without attaching RMS arm, the payload goes directly into space), 2-3 handover is done by the RMS arm code, 3-2 handover by the proximity manager code which checks a grab signal for all objects it maintains in the vicinity, 3-4 handover isn't coded but relatively straightforward (you could alter the orbit of an object in the near zone and then release into the far zone, for instance a reboost maneuver), 4-3 handover is done by the proximity manager as well (?), for all free-floating objects we're in principle able to get consistent tracking information whether they're in the near or in the far zone,...

So what are you missing? There's no difference between a released payload and something you find as you enter the near zone - from the description it just seems one of the two HST classes is redundant.
Thorsten
 
Posts: 11991
Joined: Mon Nov 02, 2009 8:33 am

Re: Space Shuttle - Development

Postby eatdirt » Sat Feb 27, 2021 10:23 am

There's no difference between a released payload and something you find as you enter the near zone


Perfect, that is indeed certainly answering the question.

- from the description it just seems one of the two HST classes is redundant.


Yes it is, we need to drop one, and that must be the one in rel_orbital.nas. I'll have a look to get it as an orbital target, and doing 4-3, that should be just changing a few conditionals. And this should work for any payload.
And 3-4 would be nice too.

thanks for the details.
cheers,
chris.
eatdirt
 
Posts: 829
Joined: Wed Aug 15, 2018 2:06 pm

Re: Space Shuttle - Development

Postby GinGin » Sat Feb 27, 2021 10:34 am

For the WSB thing , that entry video shows quite well the water consumption
https://www.youtube.com/watch?app=desktop&feature=youtu.be&v=DRczw5lG29A

It started at 96% prior to entry, ended around 66 % : 30 % for the whole entry of water consumed .

Image
GinGin
 
Posts: 1359
Joined: Wed Jul 05, 2017 10:41 am
Location: Paris
Callsign: Gingin

Re: Space Shuttle - Development

Postby Thorsten » Sat Feb 27, 2021 11:12 am

Yes it is, we need to drop one, and that must be the one in rel_orbital.nas. I'll have a look to get it as an orbital target, and doing 4-3, that should be just changing a few conditionals. And this should work for any payload.


Okay, I've just taken a look at this myself, and I believe there's a few things at odds, some of which we've understood, others not.

* The spelling 'hst' instead of 'HST' in the mission file prevents the proximity manager from finding a fitting manager when entering the near zone - I've just changed that

* I wrote the payload manager class to have something generic for objects which are transported and released and have no other requirements as a grab point. For objects with which we interact in more complex ways, I've envisioned that we have custom managers like for ISS (which has a docking point and could get multiple docking points) - for HST there was talk about servicing missions etc, so naturally I assumed we need custom code

* Wayne however rigged HST as a payload to be deployed from the bay and instanced a class under the same name, leading to ill-defined function calls

The part so far I've understood, now there's things I do not yet understand

* I can spawn either the original (O) or the payload-derived (P) hst_manager using the scenario framework. The P object is without solar panels as its animation logic expects a release, the O-object has solar panels though probably no grab point.

* Neither object registers properly with the proximity manager to provide tracking data though - this is puzzling as it works for the ISS object which I can simply spawn in addition or for the Spartan-201, so I'm not quite sure, something seems missing in there.

The way forward seems to be

a) either use a P-object and extend it to interact more than grabbing and to fix the animation logic to show deployed solar panels or
b) repair the O-object to provide a grab point somehow

It kind of depends on what kind of interactions we envision - probably a derived and extended class can go reasonably far, but for complete servicing mission it's probably easier to use a dedicated manager. Since HST is not my baby, I have no strong opinion on the point...
Thorsten
 
Posts: 11991
Joined: Mon Nov 02, 2009 8:33 am

Re: Space Shuttle - Development

Postby eatdirt » Sat Feb 27, 2021 2:52 pm

Neither object registers properly with the proximity manager to provide tracking data though - this is puzzling as it works for the ISS object


I've understood this part, or at least, made it work for the payload HST manager (the triggering is on index, so adding the index=3 works.) I am still slowly putting the pieces together, but I'll have a look to options a) and b).
eatdirt
 
Posts: 829
Joined: Wed Aug 15, 2018 2:06 pm

Re: Space Shuttle - Development

Postby wlbragg » Sat Feb 27, 2021 3:22 pm

I didn't implement 'HST the payload', so I can say very little about the specifics, but

There's four different situations that need to be addressed:

1) Object secured in the payload bay - that's a child model of the Shuttle that's simply usually not shown and a mass point definition at the right place.
2) Object attached to the RMS arm - that's a second child model coming with the relevant set of translations/rotations to follow the arm
3) Object in the near zone - that's a complicated Nasal class handling motion, attitude, contact, tracking information and interaction with the Shuttle (docking, grabbing)
4) Object in the far zone - that's a simple Nasal class providing tracking position and tracking info.


As I remember, I modeled the payload bay version of HST after Spartan-201 and so assume only situations 1) and 2) are addressed. Although 3) is pseudo implemented as there is motion handling at least, but I guess that is coming from any child model coding that is in place VS the Nasal class handling tracking as well as motion. I also didn't realize Spartan-201 could be rendezvoused with and captured, I thought ISS was the only object that had that capability. Had I know I would have looked at it a little more and possibly programmed that portion as well.
The solar cell deploy I think was accomplished by extending the HST class only. I could be wrong about that, but I think I added a property or two for that portion.
I was not aware at the time there was any other HST model or class hanging around.
Kansas(2-27-15)/Ohio/Midwest scenery development.
KEQA (2-27-15), 3AU, KRCP Airport Layout
Intel i5 3570K AMDRX480
wlbragg
 
Posts: 6006
Joined: Sat Aug 25, 2012 11:31 pm
Location: Kansas (Tornado Alley), USA
Callsign: WC2020
Version: next
OS: Win10/Linux/AMDRX480

Re: Space Shuttle - Development

Postby Thorsten » Sat Feb 27, 2021 3:49 pm

the triggering is on index, so adding the index=3 works.)


Ah, good - one more thing solved...

The solar cell deploy I think was accomplished by extending the HST class only. I could be wrong about that, but I think I added a property or two for that portion.


We might want a general deployment mechanism... Like a trigger inside the class which starts a function to fully deploy a satellite (the TDRS in its proper form would also have solar cells, the current version is really my mutilation of the 3d model...) after a specified time or so.

The HST select animation seems to query distance, payload selection and release event, but we should use something inside the private property space of the object I think.
Thorsten
 
Posts: 11991
Joined: Mon Nov 02, 2009 8:33 am

PreviousNext

Return to Spaceflight

Who is online

Users browsing this forum: No registered users and 1 guest