Jump to content

Boris-Barboris

Members
 View Profile  See their activity

  • Posts

    761

  • Joined

  • Last visited

 Content Type 

Everything posted by Boris-Barboris

  1. Boris-Barboris
    1). Use larger aoa range, 30+ will show the whole picture better. 2). Simple rule - center of mass of your fuel should be as close as possible to CoM of the craft completely dry. Only then you'll be able to forget about CoM shift due to fuel burnout (only to discover that you need to balance cargo load too). There is a mod somewhere that pumps fuel in realtime to keep CoM at the same place it was in SPH, google should help you, but it's the way of casuals. Besides, it only delays the problem, it does not solve it. Typical plane in KSP - engines are far in the back, fuel forward - nowhere near the way real planes are built. That's a rocket's composition, and rockets are unstable. Good way to balance fuel-chassis pair is to have some fuel blocks on your craft that you can easily move forward-backwards while looking at the graph (see pic below). When green and yellow line merge, you hit the right spot. After that you move wings again to balance the craft and get stability, then fuel blocks again. After a couple of iterations you'll reach good balance.
  2. Boris-Barboris
    https://github.com/Boris-Barboris/AtmosphereAutopilot/releases/tag/v1.5.9 Changelog: Version 1.2.1+ control surface module compatibility. CruiseFlight autopilot GUI tweaks by CraigCottingham. You can now enter waypoint coordinates in text form. @EpicSpaceTroll139 you had the same bug.
  3. Boris-Barboris
    Yes.
  4. Boris-Barboris
    @RobinVerhulstZ got the fix, will roll out at the evening. KSP patch shenanigans.
  5. Boris-Barboris
    http://pastebin.com/index.php or some cloud storage like dropbox\whatever.
  6. Boris-Barboris
    @RobinVerhulstZ Good job. Starting at 1:17 - the next 5 seconds (and all the next crashes), canards become inversed for some reason. Can you share the log file and the craft file? Will I need mods other than KAX to fly it?
  7. Boris-Barboris
    In case you'll fail to reproduce - it's OK, AA is inconsistent, sometimes it will crash you for no reason but will not do it when you want it to. Such is the price for ambition.
  8. Boris-Barboris
    Some visualisable example would be a good start for investigation.
  9. Boris-Barboris
    Choose for yourself. I use it everywhere, but tastes and preferences tend to differ.
  10. Boris-Barboris
    You know about stock trim keys, right? Or you means it's actually unstable (wich would mean the graph is correct)?
  11. Boris-Barboris
    @gilflo Well, second guess would be those gears. And the last one - try it without AA on 3-4% fuel tanks. After that it's clearly a bug. It will be hard to find though, since KSP does not have a proper API, and mods can also interfere. I do sometimes notice some inconsistencies between actual flight and graph, but it's just too hard to diagnose, so I usually just ignore it.
  12. Boris-Barboris
    @gilflo does the graph change, if you move the craft by it's root part a little?
  13. Boris-Barboris
    @EpicSpaceTroll139 got it twice, very inconsistent, but both times those elevons clipped into the kokpit behaved strangely. I suggest turning off roll and yaw on them.
  14. Boris-Barboris
    @EpicSpaceTroll139 Not known, let me know if you can find the reproduction craft. And, if you find it, try it without AA activated, maybe I've made some strange mistake in control surface code.
  15. Boris-Barboris
    If you would examine the function prototype, you would see that the way you do it assigns upVector to view. It is not what you need.
  16. Boris-Barboris
    https://docs.unity3d.com/ScriptReference/Quaternion.SetLookRotation.html current position - planet position = up vector view vector doesn't really matter for rocket case, pick any orthogonal to up vector.
  17. Boris-Barboris
    It always backfires. I don't think there is any risk of updates breaking something. The only thing there is to brake is jet throttle dynamics, which stayed the same as long as i can remember. And it's not like it's hard to fix after update. NN is a capricious, unreliable, untunable overkill, wich will require well-crafted sandbox anyways, and a big pile of crutches on top of it. At least for attitude. Higher order stuff, especially hybrids, may actually be helpful, but I have yet to come up with anything good.
  18. Boris-Barboris
    I'd recommend wasting a week and writing yourself a sandbox. Simple matlab\watever script is faster to work with than KSP. Implement one axis of rotation, write out a couple of cases you want to check your controller on (small, medium, heavy craft, jets or rockets), look at the plot, tweak, iterate. Trial and error is the way to go, the trick is to minimize iteration count, period and environment complexity. On the controller itself, well, PIDs are fine, the easiest and well-tested approach is gain scheduling - instead of woodoo coefficients you interpolate a big table of gains, tested for each test craft (will get trickier if table dimension is high, but your case seems to have one dimension - ratio of max torque to axis MOI). You can make two tables - for jets and for rockets. Then maybe some global gain around 0.5 to ease the fact that you omitted aero forces (don't even go there, trust me on this). I've seen a formula somewhere in RPC thread for analythical PI gains from desired overshoot margin, but I don't think it will help you - you need PID, unless you're cascading and already have good angular acceleration controller.
  19. Boris-Barboris
    Most probably tweakscale applying different exponents to different parameters - thrust squared and mass power 2.5 for example. I wouldn't assume it's the same craft mathematically.
  20. Boris-Barboris
    Ah, if only i had more time.
  21. Boris-Barboris
    Nobody expects the Spanish Inquisition!
  22. Boris-Barboris
    1. Lowest and precisely as needed Pe means you won't need additional burns to correct you trajectory in the middle of the transfer -> can mean fuel efficiency. It is only optimal when you define optimal as "get lowest Pe". Task of getting optimal ejection when optimal is defined like "get this particular Kerbin Pe from this particular Mun orbit while wasting minimum fuel" is more interesting. 2. It's a rule of thumb, and a convenient one, but it is not optimal. 3. No: different problem. Neither there is any relation to Hohmann transfer. Most of optimisation tools you'll find for KSP handle planet-planet hops, because they are the ones that usually matter. People rarely care about spare 10m\s when playing with Kerbin moons. I'm not even sure that KSP TOT cares. I'd recommend to just play the game the way intuition tells you to, it's more fun anyways. Word 'optimal' is mentioned far too often for my taste on this forum.
  23. Boris-Barboris
    @brainy try https://dl.dropboxusercontent.com/u/23572905/KSP_Mods/AtmosphereAutopilot/AtmosphereAutopilot_1.5.9.zip
  24. Boris-Barboris
    Version file? Pull request is prefered method of course, 2lazy.
  25. Boris-Barboris
    https://github.com/Boris-Barboris/CorrectCoL/releases/tag/v1.4.4
×
×
  • Create New...