Jump to content

Intersect markers disappearing


Recommended Posts

On 7/15/2016 at 5:38 AM, taniwha said:

Another thing: the point of actual closest approach between the vessel and the target (as opposed to the target's orbit) is of no interest: it does not tell us how to get closer. Showing this closest approach is one mistake that KSP currently makes for celestial body targets, and it does cause a lot of confusion.

Thank you for saying this. I've thought it for years. :)

Link to comment
Share on other sites

On 7/15/2016 at 2:38 AM, taniwha said:

Another thing: the point of actual closest approach between the vessel and the target (as opposed to the target's orbit) is of no interest: it does not tell us how to get closer. Showing this closest approach is one mistake that KSP currently makes for celestial body targets, and it does cause a lot of confusion.

31 minutes ago, 5thHorseman said:

Thank you for saying this. I've thought it for years. :)

Why have you thought it for years? 

Ultimately, the closest approach between vessel and target determines if there will be an SoI change.    So under the hood, KSP has to check for closest distances between bodies.   The positions at closest approach between bodies gives me as a user a preview of how the craft and body will be oriented (gravity assist to increase or decrease energy, for example) as I adjust the orbits.  

The closest approach of orbits, on the other hand, tells me most obviously that I wish I could follow this same path just a bit earlier or later in time, but there is not much I can do to adjust timing without changing the path.

Do you mean that future orbits might have different timing, and that the closest approach of the paths shows us the ultimate potential for a very close encounter between the bodies, maybe some orbits in the future?

Do you mean we could use the magenta and orange intersect markers for encountering bodies with noticeable gravity, just as we use them for rendezvous ?

Edited by OHara
Link to comment
Share on other sites

3 minutes ago, OHara said:

Why have you thought it for years? 

Because it's a Truth? :)

The very first time I tried to get to Gilly - somewhere around 2.5-3 years ago - I couldn't understand why the "close approach" markers were showing where Gilly was when my ship was some seemingly random place in my orbit, and not what was obviously wanted which is where Gilly will be when my ship is closest to Gilly's orbit.

This is so obvious to me it's like wanting rockets to go up when you thrust down. If someone wants something else, they should have to check or uncheck a settings box.

EDIT: Sure KSP needs to check for the other closest approach to work under the hood and sure it can show this marker if it wants to, but it's superfluous to my gameplay and at BEST should only be shown in ADDITION to the markers that are actually useful.

6 minutes ago, OHara said:

Do you mean that future orbits might have different timing, and that the closest approach of the paths shows us the ultimate potential for a very close encounter between the bodies, maybe some orbits in the future?

Do you mean we could use the magenta and orange intersect markers for encountering bodies with noticeable gravity, just as we use them for rendezvous ?

I mean both of these things, yes. Also, I mean that the markers should be consistent between non-gravitational and gravitational bodies.

Edited by 5thHorseman
Link to comment
Share on other sites

7 minutes ago, 5thHorseman said:

I couldn't understand why the "close approach" markers were showing where Gilly was when my ship was some seemingly random place in my orbit, and not what was obviously wanted which is where Gilly will be when my ship is closest to Gilly's orbit.

I suppose this is a case where you can change your timing without changing the path very much, by orbiting Kerbin or timing your launch.  It is also a case more like rendezvous, because the target SoI is so small.  The orange and magenta markers (when stable) would work well for that kind of encounter.

The orange and magenta intersect markers would be useful for encounters like the Earth gravity assist in the recent Juno mission to Jupiter, where the assist came from the second crossing of Earth's path after its maneuver.  Right now, KSP marks that second crossing if the timing is close-enough-to-desired that the bodies are closest at the second crossing. 

(On the topic of debugging, though, under the hood KSP had better have code to check that seemingly random place in the orbit where the bodies are actually closest, lest it miss an SoI change.)

Link to comment
Share on other sites

On 18/07/2016 at 6:31 AM, OHara said:

On the topic of debugging, though, under the hood KSP had better have code to check that seemingly random place in the orbit where the bodies are actually closest, lest it miss an SoI change.

By my thinking, even for that, you don't want actual closest approach, at least not directly (though it looks like KSP does currently have something like that). Instead, find the ranges in which the vessel is in the SoI corridor of the body's orbit, and then see if the body is within SoI range of the vessel's orbit during the time span of those ranges. If so, then refine to find the earliest time the vessel touches the body's SoI, assuming it does (they may still miss each other).

However, this is a later project that depends on the improved targeting code, as there are quite a few issues with the patched conics quite separate to the intercept issues such as your orbit encounters both Minmus and Mun, but Mun is after Minmus: KSP shows the Mun encounter, not the Minmus encounter). I've had that happen, I was not amused.

Link to comment
Share on other sites

Every time I play KSP now I notice more about intersections.  I'll just say what I notice and trust taniwha to ignore anything that's only distraction.

On 7/15/2016 at 2:38 AM, taniwha said:

Another thing: the point of actual closest approach between the vessel and the target (as opposed to the target's orbit) is of no interest: it does not tell us how to get closer.

In some (possibly trivial) cases the closest approach tells us how to get closer: mid-course corrections.  Point in the direction from cyan marker indicating the vessel to that for the target, and burn a dV of   closest-approach-distance / time-until-closest-approach.

That is the minimum-dV burn from first-order perturbation-theory, so it is inaccurate to the extent that changing the vessel path changes the gravitational field that the vessel flies through, so it is useful when there is less than about 1/4 orbit around anything between me and my goal.

On 7/15/2016 at 9:49 PM, taniwha said:

This [rotating] does not work because when you rotate an orbit into another plane, you change the location of its foci. Take the obvious example: rotate a circle: it's one focus (the center) splits in two and the diverge along the AN/DN axis.

I meant literally rotating the vessel orbit in 3D space, with no change in its shape, leaving one focus on the sun (or whatever body we are orbiting) as one would do with a burn perpendicular to the orbital plane.  Then there are no more than two intersections and the solution is easy (thanks to the nice people who developed algorithms for arcsine and such).

I agree that projecting orbits onto the same plane after rotation can move the foci as you say, open up the possibility of four intersections, and I see no analytic solution for those intersections.  

On 7/19/2016 at 2:18 AM, taniwha said:

By my thinking, even for that [finding SoI changes], you don't want actual closest approach [between orbital paths], at least not directly (though it looks like KSP does currently have something like that). Instead, find the ranges in which the vessel is in the SoI corridor of the body's orbit,

Those ranges, I guess, are in the two-dimensional space comprising pairs of positions, one along each of the two orbits.  Some but not all the closest approaches (*s in my figure) will be in those ranges.  I plotted the curves where each orbit's tangent is perpendicular to the separation of points on the two orbits (the other condition you mentioned) and see 9 places where both tangnts are perpendicular to the separation, three of which are local closest approaches.

inclineOrbits.png

I was thinking that the SoI search would be a 1-D search of positions as a function of time.  It is possible that the code tried to avoid computing a lot of orbital positions at given times (inverting the Kepler equation) in favor of finding the positions first, and then working out the times (the easier forward Kepler equation).   

I guess you have BSP sorting through the pairs of positions on the two orbits to find the subset that are within the SoI radius of each other.  I don't see how that can be extended to find all (up to four) local minima, but I'm pretty sure that out of those (up to four) minima we want to see the ones that correspond to potential (after doing things like normal burns) actual intersections.

Link to comment
Share on other sites

  • 1 month later...
On 7/16/2016 at 6:19 PM, taniwha said:

Well, I very much hope that I will not disappoint you.

It looks nice in 1.2 pre-release.  It looks like the solution for "which two intercepts to plot, when there are three or more minimum distances?", is simply the to plot the first two following the craft's final maneuver-node.  That's nice and simple, and adjustable by placing a dummy maneuver node if needed.

I like that the two near-intersections with this asteroid orbit are marked, even in the presence of significant inclination.  The periapses of the craft and target orbits is a closer local-minimum distance between the orbital paths, but not a useful rendezvous point.  As long as my craft's maneuver is after her periapsis, KSP plots the two useful close-approaches in deeper space.

Spoiler

Inclination.jpg

Plotting later close-encounters and repeated SoI entries is nice as well.

Spoiler

nearSecondEncounter.jpgsecondEncounter.jpg

 

Edited by OHara
Link to comment
Share on other sites

  • 2 years later...
  • 1 year later...
This thread is quite old. Please consider starting a new thread rather than reviving this one.

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...