Board index FlightGear Development Spaceflight

Space Shuttle - Development

Discussion about development and usage of spacecraft

Re: Space Shuttle - Development

Postby eatdirt » Sun Mar 28, 2021 12:20 pm

Thanks Wayne!
I am currently trying to see if everything goes fine in copy_offsets(), that's the place where the 3D model gets its location after release. The fact that the model motion is fine at grabbing but jump at released is weird, may the problem is not at all animation related?
eatdirt
 
Posts: 833
Joined: Wed Aug 15, 2018 2:06 pm

Re: Space Shuttle - Development

Postby Thorsten » Sun Mar 28, 2021 12:57 pm

I am currently trying to see if everything goes fine in copy_offsets(), that's the place where the 3D model gets its location after release.


Yes, everything goes fine in there. I've been releasing three HST and two Spartans over the last weeks, every time the function worked as advertized. If something works before you add animations and ceases to work afterwards, then logic suggests it is animation-related.

Also, it seems to me you're working the wrong way. We have a working code that transfers HST from free-float consistently to the RMS arm and back, including all the rotations/twists you like.

The only thing that's not working for you is the transfer to/from the stowed position - which suggest it's easiest to change that, because it generally is harmless in its interaction with anything else. You seem to be trying to do the opposite - which is a recipe for disaster as changes there are not harmless.
Thorsten
 
Posts: 12000
Joined: Mon Nov 02, 2009 8:33 am

Re: Space Shuttle - Development

Postby eatdirt » Sun Mar 28, 2021 1:28 pm

Yes, everything goes fine in there. I've been releasing three HST and two Spartans over the last weeks, every time the function worked as advertized.


Well, at least your sentence made me just tried that. It does not :) With current devel snapshot, take Hubble out of the bay, when you release it, it jumps by the amount I am describing. I was actually quite stupid to not check this first :-/

Edit: better in image, this is the whole sequence, because if you don't see that, it means that the jump might be induced by a chain of events on the arm: test_unload_HST_vp6.webm

If something works before you add animations and ceases to work afterwards, then logic suggests it is animation-related.


Indeed. So, I would say everything is good for the merge request.

The only thing that's not working for you is the transfer to/from the stowed position


I may have been unclear, but that is working fine! In fine, the only problem I am seeing is not induced by my changes!
eatdirt
 
Posts: 833
Joined: Wed Aug 15, 2018 2:06 pm

Re: Space Shuttle - Development

Postby Thorsten » Sun Mar 28, 2021 5:47 pm

With current devel snapshot, take Hubble out of the bay, when you release it, it jumps by the amount I am describing.


I've just taken another 15 minutes to lift the HST out of the bay, lift it up, rotate it somewhat for a challenge, take a screenshot, release it, take another screenshot, and I see no shift in either position or attitude - the screenshots are visually identical.

So whatever fault exists, it's not reproducible on my side, my test in detail confirms what I've seen before without screenshot comparison - the release code works fine. Whatever fault is seen on your side is probably caused by a change you did (or possibly by the latest changes Wayne did with the grappling fixture- I've not copied them over yet).
Thorsten
 
Posts: 12000
Joined: Mon Nov 02, 2009 8:33 am

Re: Space Shuttle - Development

Postby wlbragg » Sun Mar 28, 2021 6:36 pm

The target rotation should not change anything as long as the center didn't change. The wrist roll rotation maybe, but I would think that would only pertain to payload roll? Except all the arm rotations would possibly change everything.
Kansas(2-27-15)/Ohio/Midwest scenery development.
KEQA (2-27-15), 3AU, KRCP Airport Layout
Intel i5 3570K AMDRX480
wlbragg
 
Posts: 6012
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 eatdirt » Sun Mar 28, 2021 8:55 pm

Whatever fault is seen on your side is probably caused by a change you did (or possibly by the latest changes Wayne did with the grappling fixture- I've not copied them over yet).


I've just done a complete delete of .Aircraft/SpaceShuttle as well as of my git copy of it, recloned everything from the official fgspaceshuttle-code repo, started from scratch with stable release of FG (2020.3.6), with latest commit of fgspaceshuttle-code on the devel branch.

I am starting with this:

fgfs --fg-aircraft=/home/chris/.Aircraft --aircraft=SpaceShuttle-mission --prop:/sim/config/shuttle/mdu-groups=1


which starts above BIKF. No change, HST jumps at release time. Next test, I reverted the commit from Wayne: no change, HST still jumping!

It would be nice if someone else could test this, or explain to me what remains to be tested that could explain why I do see this and not Thorsten.

Edit: Mmm, I've made a test starting fgfs with --lat=0 and --lon=0, the main jump on y is completely gone, I do see only a little one remaining on the x axis: @Thorsten, at which --lat --lon do you release?
That smells the subtle bug to me, or some proximity coordinates accuracy issues?
eatdirt
 
Posts: 833
Joined: Wed Aug 15, 2018 2:06 pm

Re: Space Shuttle - Development

Postby Thorsten » Mon Mar 29, 2021 5:44 am

@Thorsten, at which --lat --lon do you release?


HST tests at < 28 deg. The Spartan I've released over Antarctica though and not seen anything unusual either.
Thorsten
 
Posts: 12000
Joined: Mon Nov 02, 2009 8:33 am

Re: Space Shuttle - Development

Postby eatdirt » Mon Mar 29, 2021 9:56 pm

HST tests at < 28 deg. The Spartan I've released over Antarctica though and not seen anything unusual either.


I am narrowing down the options, my best guess is FG level bug, possibly some axis convention. I see problems on "y" only, proportional to sin(latitude). And as I posted a while ago, when going to ISS, I've observed reversed Ydot on RDV spec. Are you using 2018.* ? If yes, which version exactly?
eatdirt
 
Posts: 833
Joined: Wed Aug 15, 2018 2:06 pm

Re: Space Shuttle - Development

Postby GinGin » Tue Apr 06, 2021 11:36 am

What is left to test and improve before a stable release ?
GinGin
 
Posts: 1364
Joined: Wed Jul 05, 2017 10:41 am
Location: Paris
Callsign: Gingin

Re: Space Shuttle - Development

Postby Thorsten » Tue Apr 06, 2021 4:56 pm

Not much I guess. I wanted to do some rendezvous tests still, but that's about it from my side.

Who is handling the RMS cam view merge request? Is everyone waiting for me to do that, or is someone already familiar with it available?
Thorsten
 
Posts: 12000
Joined: Mon Nov 02, 2009 8:33 am

Re: Space Shuttle - Development

Postby eatdirt » Tue Apr 06, 2021 6:54 pm

Who is handling the RMS cam view merge request? Is everyone waiting for me to do that, or is someone already familiar with it available?


It should be up to date with latest devel commits, I've refreshed it on regular basis. My only concern would be that Wayne should have a look to the new hst.ac, everything looks fine to me, but I am very new to blender!!
eatdirt
 
Posts: 833
Joined: Wed Aug 15, 2018 2:06 pm

Re: Space Shuttle - Development

Postby GinGin » Wed Apr 07, 2021 11:55 am

Who is handling the RMS cam view merge request?


I don’t have a computer until next week. I tested the RMS view during polar mission and it worked nicely.
I wanted to to do at least a full rdv also, and I have the logic you suggested for the mission file launch location check.


All the rest I am working at ( Mainly TAEM and Entry deeper logic) will be for post stable dev as everything we tested seems to work nicely now and without major flaws.
GinGin
 
Posts: 1364
Joined: Wed Jul 05, 2017 10:41 am
Location: Paris
Callsign: Gingin

Re: Space Shuttle - Development

Postby Thorsten » Wed Apr 07, 2021 1:00 pm

@GinGin, @wlbragg:

Okay, if you have tested the RMS and Wayne knows the 3d model issues - could you handle the RMS view commit? I'd have to work my way in fron scratch...

I just tested a final rendezvous phase with the Shuttle bring targeted some 10 km ahead of ISS and then going back until docking - this basically worked okay, including changing from 2 engine to 1 engine OMS burn.

The only thing is that the Lambert solver can to exit when the solution is too complicated to find in a way that blocks further calls of it permanently. So I guess I need to insert a manual shutdown flag when item 28 is selected again while the computation flag is still on that terminates the current computation and re-sets the solver, as we have no other way of doing that...

That would leave:

* the RMS view commit review
* the LOW ENERGY CWS warnings blocked when the Shuttle is on the carrier aircraft
* a look into the Contingency engage logic to see what causes the acceleration spikes
* the Lambert issue outlined above

... and that should cover the remaining issues.

So I'm basically done with what I wanted to test, and will focus more on the manual now. 8)
Thorsten
 
Posts: 12000
Joined: Mon Nov 02, 2009 8:33 am

Re: Space Shuttle - Development

Postby Vinny002 » Wed Apr 07, 2021 8:11 pm

I am very excited about trying out the stable version of the FG space shuttle!! Cheers vincent.
Vinny002
 
Posts: 73
Joined: Mon Jul 01, 2019 8:55 pm

Re: Space Shuttle - Development

Postby wlbragg » Wed Apr 07, 2021 9:11 pm

So where is the new hst.ac file to review, in the development branch or somewhere else. The only branch or location I ever have accessed is the development branch using a local GIT GUI?
Kansas(2-27-15)/Ohio/Midwest scenery development.
KEQA (2-27-15), 3AU, KRCP Airport Layout
Intel i5 3570K AMDRX480
wlbragg
 
Posts: 6012
Joined: Sat Aug 25, 2012 11:31 pm
Location: Kansas (Tornado Alley), USA
Callsign: WC2020
Version: next
OS: Win10/Linux/AMDRX480

PreviousNext

Return to Spaceflight

Who is online

Users browsing this forum: No registered users and 1 guest