Youen

[old thread] Trajectories : atmospheric predictions

Recommended Posts

does this work with heatshields/fairings in stock?

I will test a fairing-missile probe right now and see how accurately I can land it without ejecting the fairing... be right back with results.

Vehicle:

Gxxpfi3.jpg?2

Attempt #1

hNim23d.jpg

Long, flat trajectory with the missile pointed prograde through the atmosphere.

Result

axNEmEk.jpg

Passed over KSC at 31.8 km. Image is darker than I wanted, the impact point is just off the coast of the next continent.

It seems that long atmospheric flights tend to have the most inaccurate predictions, but this is true for all vessel types. The longer and flatter your trajectory, the more mistakes in the prediction and the less accurate it is.

Attempt #2

FWyXBEu.jpg

More aggressive, steeper trajectory.

Result

Passed over KSC at 5.6 km, impacted about 11.5 km from the VAB (where my flag is). I forgot to get a screenshot before I attempted to refocus the worldmap and lost the impact point.

TL;DR - Yes, it works with fairing and heat shields. The more aerodynamic the vessel, the more inaccurate the prediction (it seems). It may just be worth it to make a test landing and see how far off your vessel is from the prediction, then correct for it. Generally speaking, aim short.

Edited by Baythan

Share this post


Link to post
Share on other sites

Is there a way to see the exact predicted keolocation (lat long) of predicted landings, using FAR? I'm trying precision landings to within 5 meter radii. With stock, Mechjeb works ok with a little manual steering during descent.

Share this post


Link to post
Share on other sites

Could the first post please be updated? The title says “FAR onlyâ€Â. The FAQ says “stock aerodynamics and FARâ€Â, which appears to be correct. The Mod Description section says stock only. Kerbal Stuff says Stock, FAR and NEAR.

Share this post


Link to post
Share on other sites
Could the first post please be updated? The title says “FAR onlyâ€Â. The FAQ says “stock aerodynamics and FARâ€Â, which appears to be correct. The Mod Description section says stock only. Kerbal Stuff says Stock, FAR and NEAR.

As noted 2 pages ago:

I would have, but it's not my thread so in fact I can't. I've just PM'd Kobymaru but no guarantee he'll be arround. Feel free to subscribe to the notifications on KerbalStuff though (at least there we can add multiple developpers to the same mod so anyone can publish updates).

Share this post


Link to post
Share on other sites

Please update the title to remove the FAR only or mention that it does work with stock as I never noticed until I read the description. IT would bring more people to the download!

Share this post


Link to post
Share on other sites

Will this ever be useable stock again, or are the new aero properties too unpredictable?

EDIT:

Please update the title to remove the FAR only or mention that it does work with stock as I never noticed until I read the description. IT would bring more people to the download!

:blush:

Share this post


Link to post
Share on other sites

If this is being taken over by a different modder, maybe it's time to move to a new thread so he can edit the op, unless he is just filling in for awhile.

Share this post


Link to post
Share on other sites

AFAIK, there are several folks still working on it, but with varying amounts of time available.

Share this post


Link to post
Share on other sites
AFAIK, there are several folks still working on it, but with varying amounts of time available.

What he said...

Share this post


Link to post
Share on other sites

If someone else is maintaining the mod, they should start a new thread. Not having control of a thread is bad for a modder.

Good to know it works now. I (like I suspect many) assumed the title was correct.

Share this post


Link to post
Share on other sites

Actually, the original thread was mine (http://forum.kerbalspaceprogram.com/threads/93685), and we switched here when Kobymaru started to work on the port for 0.90, because I had no time to work on the mod at this time. But now it's the opposite. Maybe the best solution would be to create a new account (and a new thread) so that we can share the password for all developpers.

Share this post


Link to post
Share on other sites
Actually, the original thread was mine (http://forum.kerbalspaceprogram.com/threads/93685), and we switched here when Kobymaru started to work on the port for 0.90, because I had no time to work on the mod at this time. But now it's the opposite. Maybe the best solution would be to create a new account (and a new thread) so that we can share the password for all developpers.

Actually, shared accounts aren't allowed on this forum. We moderators will actively ban any we find.

But, as to help with management of threads started by a different author, that's something we do quite often. Just PM me or any other moderator you like, we will make possible either to change content in the Opening Post or even to switch thread ownership.

Share this post


Link to post
Share on other sites
Could the first post please be updated? The title says “FAR onlyâ€Â. The FAQ says “stock aerodynamics and FARâ€Â, which appears to be correct. The Mod Description section says stock only. Kerbal Stuff says Stock, FAR and NEAR.
Please update the title to remove the FAR only or mention that it does work with stock as I never noticed until I read the description. IT would bring more people to the download!

It has been updated (a few days ago?). NEAR does not exist for 1.0.2, or am I mistaken? Has anyone tested this for NEAR?

If this is being taken over by a different modder, maybe it's time to move to a new thread so he can edit the op, unless he is just filling in for awhile.
If someone else is maintaining the mod, they should start a new thread. Not having control of a thread is bad for a modder.

Good to know it works now. I (like I suspect many) assumed the title was correct.

Sure, Youen can have his thread back if he wants. But this is exactly what we did when we thought that I would have more time to work on it than him. Might switch it back, might keep it.

Actually, shared accounts aren't allowed on this forum. We moderators will actively ban any we find.

But, as to help with management of threads started by a different author, that's something we do quite often. Just PM me or any other moderator you like, we will make possible either to change content in the Opening Post or even to switch thread ownership.

Yeah, we can change thread ownership back to Youen, if he wants. If anyone has noticed, I updated the first post multiple times to give maximum credit to actual authors (can't forward reputation though) and changed all the links to Youens Github page instead of mine.

It's your call! PM me and a Mod if you want to see this happen.

Edited by Kobymaru

Share this post


Link to post
Share on other sites

Well many thanks from me to Youen, Kobymaru, and atomicfury for all your contributions to the mod. Its a very nice one to have and the new aero makes it even more important!

Share this post


Link to post
Share on other sites

I've got v1.3, which claims to work with stock air. And I'm using stock air, not FAR. However, the predictions I'm getting for Ap after aerocapture aren't even close until after my ship has passed Pe, which makes them essentially useless. It goes like this:

1. As the ship approaches the planet, I adjust my Pe down into the atmosphere and Trajectories shows a prediction. This prediction appears to be based on the old pre-1.x atmosphere data. That is, for a given Pe, it shows the resulting Ap just like how things were back in the day. For example, coming into Duna, Trajectories is saying a Pe between 11-12km is what you want. However, in 1.0.2, that will lithobrake you---what you really want these days is 16-18km.

2. As the ship enters the atmosphere, the predicted path doesn't change to start with. However, as the ship reaches Pe, suddenly the predicted path jumps to a much higher final Ap, and this turns out to be about where the ship eventually ends up after exiting the atmosphere.

The mod seems to be able to figure out the stock aero OK, given what it shows after Pe. But it's way off prior to Pe, which rather defeats the whole purpose of the mod. I hope this is just some sort of display issue, not a calculation issue :). Either way, though, the mod is failing at predicting where you'll end up before you enter the atmosphere, which means I'm still have to do the old F5, see what happens, F9, tweak Pe, F5, repeat process even with Trajectories installed.

Share this post


Link to post
Share on other sites

Since I updated Trajectories, I've used it mostly for landing on Kerbin. It's definitely not perfect, but its also better than nothing. Overall, the landing prediction is somewhere between being good and my actual landing position overshooting the initial spot. However, even then there appears to be 3 phases. 1) In the upper atmosphere actual overshoots the prediction, 2) in the middle of the atmosphere during the deceleration and heating phase my actual position then switches to undershooting the prediction, and 3) In the lower atmosphere its more accurate, but still has a tiny tendency for actual to overshoot the prediction. 1) and 2) almost cancel out for landing, but I can see that aerobraking probably gets more of phase 1 and overshoots more.

Share this post


Link to post
Share on other sites

Got a crash.

Output_log.txt is flooded with that:

NullReferenceException

at (wrapper managed-to-native) UnityEngine.Transform:INTERNAL_get_worldToLocalMatrix (UnityEngine.Matrix4x4&)

at UnityEngine.Transform.get_worldToLocalMatrix () [0x00000] in <filename unknown>:0

at FerramAerospaceResearch.FARAeroComponents.FARAeroSection.PredictionCalculateAeroForces (Single atmDensity, Single machNumber, Single reynoldsPerUnitLength, Single skinFrictionDrag, Vector3 vel, ferram4.FARCenterQuery center) [0x00000] in <filename unknown>:0

at FerramAerospaceResearch.FARAeroComponents.FARVesselAero.SimulateAeroProperties (UnityEngine.Vector3& aeroForce, UnityEngine.Vector3& aeroTorque, Vector3 velocityWorldVector, Double altitude) [0x00000] in <filename unknown>:0

at FerramAerospaceResearch.FARAPI.InstanceCalcVesselAeroForces (.Vessel vessel, UnityEngine.Vector3& aeroForce, UnityEngine.Vector3& aeroTorque, Vector3 velocityWorldVector, Double altitude) [0x00000] in <filename unknown>:0

at FerramAerospaceResearch.FARAPI.CalculateVesselAeroForces (.Vessel vessel, UnityEngine.Vector3& aeroForce, UnityEngine.Vector3& aeroTorque, Vector3 velocityWorldVector, Double altitude) [0x00000] in <filename unknown>:0

at (wrapper managed-to-native) System.Reflection.MonoMethod:InternalInvoke (object,object[],System.Exception&)

at System.Reflection.MonoMethod.Invoke (System.Object obj, BindingFlags invokeAttr, System.Reflection.Binder binder, System.Object[] parameters, System.Globalization.CultureInfo culture) [0x00000] in <filename unknown>:0

Rethrow as TargetInvocationException: Exception has been thrown by the target of an invocation.

at System.Reflection.MonoMethod.Invoke (System.Object obj, BindingFlags invokeAttr, System.Reflection.Binder binder, System.Object[] parameters, System.Globalization.CultureInfo culture) [0x00000] in <filename unknown>:0

at System.Reflection.MethodBase.Invoke (System.Object obj, System.Object[] parameters) [0x00000] in <filename unknown>:0

at Trajectories.VesselAerodynamicModel.computeForces_FAR (Double altitude, Vector3d airVelocity, Vector3d vup, Double angleOfAttack, Double dt) [0x00000] in <filename unknown>:0

at Trajectories.VesselAerodynamicModel.computeFARReferenceDrag () [0x00000] in <filename unknown>:0

at Trajectories.VesselAerodynamicModel.isValidFor (.Vessel vessel, .CelestialBody body) [0x00000] in <filename unknown>:0

at Trajectories.Trajectory+<computeTrajectoryIncrement>d__7.MoveNext () [0x00000] in <filename unknown>:0

at Trajectories.Trajectory.ComputeTrajectory (.Vessel vessel, Trajectories.DescentProfile profile, Boolean incremental) [0x00000] in <filename unknown>:0

Unity Player [version: Unity 4.6.4f1_99f88340878d]

KSP.exe caused an Access Violation (0xc0000005)

in module KSP.exe at 0023:0127dbff.

Error occurred at 2015-06-07_124134.

67% memory in use.

0 MB physical memory [2653 MB free].

0 MB paging file [0 MB free].

0 MB user address space [123 MB free].

Write to location 00000008 caused an access violation.

Share this post


Link to post
Share on other sites
I've got v1.3, which claims to work with stock air. And I'm using stock air, not FAR. However, the predictions I'm getting for Ap after aerocapture aren't even close until after my ship has passed Pe, which makes them essentially useless.
Overall, the landing prediction is somewhere between being good and my actual landing position overshooting the initial spot.

Just to make sure, are you staging off any parts of your craft during this sequence of events?

Share this post


Link to post
Share on other sites
Just to make sure, are you staging off any parts of your craft during this sequence of events?

Nope, I'm not staging during aerocapture. Whatever left Kerbin's SOI enters the atmosphere of Eve/Duna/Jool/Laythe with all the same parts it left with, just less any fuel used for mid-course corrections and final Pe tweaks before hitting the air. Then, assuming I guessed the Pe correctly, it goes on to other maneuvers.

Share this post


Link to post
Share on other sites

My trajectories arent working...I have a ship coming into the sphere and it doesnt show anything idk what I am doing wrong I got 1.0.2 KSP

Share this post


Link to post
Share on other sites
I've got v1.3, which claims to work with stock air. And I'm using stock air, not FAR. However, the predictions I'm getting for Ap after aerocapture aren't even close until after my ship has passed Pe, which makes them essentially useless. It goes like this:

1. As the ship approaches the planet, I adjust my Pe down into the atmosphere and Trajectories shows a prediction. This prediction appears to be based on the old pre-1.x atmosphere data. That is, for a given Pe, it shows the resulting Ap just like how things were back in the day. For example, coming into Duna, Trajectories is saying a Pe between 11-12km is what you want. However, in 1.0.2, that will lithobrake you---what you really want these days is 16-18km.

2. As the ship enters the atmosphere, the predicted path doesn't change to start with. However, as the ship reaches Pe, suddenly the predicted path jumps to a much higher final Ap, and this turns out to be about where the ship eventually ends up after exiting the atmosphere.

The mod seems to be able to figure out the stock aero OK, given what it shows after Pe. But it's way off prior to Pe, which rather defeats the whole purpose of the mod. I hope this is just some sort of display issue, not a calculation issue :). Either way, though, the mod is failing at predicting where you'll end up before you enter the atmosphere, which means I'm still have to do the old F5, see what happens, F9, tweak Pe, F5, repeat process even with Trajectories installed.

Does your craft have any wings, control surfaces or fairings? The stock aero prediction system isn't working quite right with those at the moment.

Share this post


Link to post
Share on other sites
Does your craft have any wings, control surfaces or fairings? The stock aero prediction system isn't working quite right with those at the moment.

Yup, fairings.

Share this post


Link to post
Share on other sites

I've been having issues using this mod to get ANY sort of accuracy on Kerbin.

I haven't been using any aerodynamic parts (wings, fairings, etc).

Only potentially relevant mod is Kerbal Joint Reinforcement.

I know of others who find the same thing.

I have profile set to retrograde and use SAS to hold this heading.

I'll run some tests if I find time and/or people want the data.

I'd love to get this mod working, I'm sure a lot of hard work went into making it, and it'd be a shame if the best thing I can find to do with the mod is use it to quickly locate the KSC.

Share this post


Link to post
Share on other sites

Can't wait for better lift/control surface support. This mod became required for me when I found it for landing space planes in beta. Plus sending oversized accent stages to KSC rod of Zeus style is fun to. Wish I had proficiencies needed to help.

I have a craft that is relatively simple (mk16, mk1 pod, service module (with occluded goodies), FL-T800, 4 RCS block, 909) that I made for some early career rescue missions. When I aerobrake with it, my trajectory is overly pessimistic (it is short of what actually happens by a significant margin). I set SAS to hold prograde and use x4 physics warp for the aerobrake. Is this due to a know bug, unknown bug, or user error?

Oh well, better to brake too little than too much.

Share this post


Link to post
Share on other sites
Guest
This topic is now closed to further replies.