Could The Gimbal Video Show an Atlas V Launch?

markus

Active Member
maybe the range comes from the ATFLIR laser range finder
If the laser was active and providing range information the indication "LTD/R" would flash on the display. Mover corroborated this, having said (without further details) that he doesn't think the laser is active.
 

Domzh

Active Member
could it be the radar reacquired the track and showing the already designated primary target distance but the atflir was independently following its own target?
 

markus

Active Member
could it be the radar reacquired the track and showing the already designated primary target distance but the atflir was independently following its own target?
My opinion is that it could, which would explain why one of the guys (presumably the pilot) was confused that the target was said to be the L&S, protesting "Well the FLIR's looking..." before being interrupted. I suspect he was about to say it was looking in a different direction than where the L&S was. But it's very hard to know for sure. For example, it could be some form of range-denying jamming/electronic warfare.
 

jarlrmai

Senior Member
If the laser was active and providing range information the indication "LTD/R" would flash on the display. Mover corroborated this, having said (without further details) that he doesn't think the laser is active.
The laser appears to be an air to ground only device for ranging and guiding laser guided air to ground munitions.
 

Mendel

Senior Member.
Trigonometry would work for ground targets
It also works for air targets, but it requires two simultaneous observations with a sufficiently wide base line (which the ATFLIR lacks) [or two non-simultaneous observations of a stationary target].
Article:
Bundesarchiv_Bild_183-J08361,_Entfernungsmesser_einer_Vierlings-Flak.jpg
The angle of sight of a rangefinder and the range to the target can be combined in a simple computer to produce a measurement of altitude. The resulting instrument becomes a combined height-finder/rangefinder and was standard equipment for land and naval based anti-aircraft units.
 
Last edited:

Edward Current

Active Member
Trigonometry would work for ground targets but it's hard to see how it could work for air targets. Using certain assumptions (e.g. target moving in a straight line) it'd be possible to automate the kind of analysis that you did in Blender, except using raw numbers instead of the rounded ones we've had to use/guesstimates from clouds. At any rate, the number pops up pretty much immediately in Gofast, but building such an estimate would require time to acquire samples as the F-18 maneuvers. Since there's ample such information in Gimbal and no range is indicated, I suspect this method isn't used. It seems more likely that the range in Gofast comes from radar.
I don't know for certain, but I think the main assumption for passive ranging is that the object isn't moving, as something on the ground would be. That's the only way such a measurement could be made. When I modeled GoFast (less precisely than I did Gimbal), there was a pivot point on the sightline where the object could be at rest. At 12,000 feet, I measured 5 knots. No such point exists in the Gimbal case; the slowest speed I could find is 75 knots around 16 NM. And in Gimbal only one angle is slewing significantly; in GoFast both are.

So, radar, maybe. But Lehto called the range number a guess, and radar wouldn't be a guess. A passive trig calculation, based on both azimuth and elevation slew rate, would be.
 

Leonardo Cuellar

Active Member
It is very interesting to understand how the Range and VC data that appear on the GOFAST video are acquired.
Reading the manual on the Hornet family Radar System, it would appear that ATFLIR can generate a trackfile, via Autotrack.
flir.png

It is evident that this trackfile if displayed on the SA page must have polar coordinates.
How it manages to get the distance I honestly have no idea, probably through triangulation or other sensor capabilities related perhaps to the spot size ratio, and it would be a great challenge to reach an answer so as not to find ourselves displaced by thorny questions.
 

Domzh

Active Member
i really dont think we should use a video game as a reliable source on how ATFLIR works. Especially how it works together with a radar.

There are multiple classified specifics that they couldnt take into account when coding the engine. They cant include rare bugs due to system failure or wear and tear.

by the way, i found an interesting study but its on request only.

anyone has a research gate account or science background to make it more likely to get the pdf from the author when requested?

google shows a preview, it seems they looked into some issues with the ATFLIR.
study was published 2006.

https://www.researchgate.net/public..._advanced_forward_looking_infrared_ATFLIR_pod
 

Attachments

  • 759AB6E8-0C1A-4915-87A7-F63C92D9C07F.jpeg
    759AB6E8-0C1A-4915-87A7-F63C92D9C07F.jpeg
    322.4 KB · Views: 91

jarlrmai

Senior Member
I have read, thanks.
Unfortunately, however, we are the usual problem that we don't know if the ATFLIR also generates AOT, as in the manual you posted it speaks only of Radar.
Yup hence the thread on this that was started to try and address Lehto's statements.

The line inquiry goes like this

  1. TTSA say Go Fast is low to the surface of the ocean.
  2. However the figures on screen (after math) how Go Fast is high, Metabunk calls this out, tracks height and speed of object based on these figures, object is shown to be high and slow, likely a balloon, probably with a RADAR reflector.
  3. Hence TTSA are wrong.
  4. Years pass, no-one really questions the specifics of that debunk (Fravor/Graves etc other people with likely experience of using the equipment)
  5. Chris Lehto releases a video about Go Fast with some wrong statements about depth of field, buried in there is the claim that the figures on the screen are wrong, inaccurate, and thus the GO FAST object can be low.

This is presented without much detail or evidence, but Lehto is a ex fighter pilot, he has some authority here so these claims must be taken seriously to some degree.

Hence this debate about where they come from.

Lehto's statement about inaccuracy is possibly intended to cast doubt on the figures which go against the low and fast theory however it really opens up a can of worms.
  • Why would the system show highly inaccurate numbers to the crew so matter of factly, no warning or anything.
  • How inaccurate can they be? If are to believe that the Go Fast object is low to the ocean then the figures are out by a few nautical miles, how is any system that can be this out be of any use?
  • Why are they inaccurate in the direction of away from the plane and not equally inaccurate in the direction of towards the plane?
  • Why did TTSA not mention these figures at all, you would think they would realise that those figures countered the low/fast theory dead, so they should have made an effort to address the reasons this in the claim?
  • There are only loose mentions of system other than RADAR that can generate distances to A/A targets.
  • If the system is capable of generating a range for Go Fast with no RADAR track then why is it not doing the same for GIMBAL.
It's of course possible there are reasonable answers to all these questions, and knowing them would enhance the understanding of what the videos show.
 

jarlrmai

Senior Member
This is the nerve center that if you want we can continue to discuss it on your thread. Do you agree?
I think all the other points are equally valid questions. I don't think you or I can get 100% to the bottom of them with the documents and information we have available, like anything else it probably needs some Raytheon engineer to tell us.

But feel free to update that thread.
 

Domzh

Active Member
a target can apparently become L&S either through radar or through atflir auto track.

we learned that once a target becomes L&S all sensors will look at it.

Do we know if the L&S on the ATFLIR is only boxed if the designation comes vom the pod itself or if it was designated by radar? or should it show L&S no matter which sensor designated it as such?

I still dont understand how the ATFLIR could designate a target L&S without range. How would the radar or missile know where to look? ATFLIR has infinite range so it could literally be everywhere within the ATFLIR LOS?
 

jarlrmai

Senior Member
a target can apparently become L&S either through radar or through atflir auto track.

we learned that once a target becomes L&S all sensors will look at it.

Do we know if the L&S on the ATFLIR is only boxed if the designation comes vom the pod itself or if it was designated by radar? or should it show L&S no matter which sensor designated it as such?

I still dont understand how the ATFLIR could designate a target L&S without range. How would the radar or missile know where to look? ATFLIR has infinite range so it could literally be everywhere within the ATFLIR LOS?

L+S is a designation for "the most important" track, a global track setting it applies to all weapons and sensors. It means Launch and Steer. It means I want to follow this target and I want to potentially launch missiles at it.

All sensors are defaulted to look at L+S for this reason, in combat I see something I want to shoot it down or follow it I want all my stuff pointed at it. These settings can be overridden, you can have and L+S and then the ATFLIR can be pointed at something else if you want. You can have no L+S and point the ATFLIR at something have it track, then make this the L+S. It all depends on your order of operation and what your sensors are showing.

Angle Only means you know the angle from your plane to the target like it's 45 degree left and 10 degree up, this doesn't tell you how far away it is but you can certainly point at it, you just don't know how far away it is and how fast it is moving.

The track system supports angle only tracks, i.e. it's literally a row in the database that has the angles from you to it but no other data.

You can imagine that if you were tracking a jamming target or a low profile one that you'd want to maintain pointing at it even if you didn't know the other information.
 

Domzh

Active Member
i still dont understand why or how ATFLIR can designate something on autotrack as L&S with its infinite optical range

wheres the difference to just tracking it because without additional information like such as distance to the target, the other sensors wouldnt know where to find it.

you know what i mean?
 

jarlrmai

Senior Member
i still dont understand why or how ATFLIR can designate something on autotrack as L&S with its infinite optical range

wheres the difference to just tracking it because without additional information like such as distance to the target, the other sensors wouldnt know where to find it.

you know what i mean?

You see something out the window, its not on RADAR
You point ATFLIR at it with a joystick on your controls
It shows up on the screen as a black/white blob as its colder/hotter than it's surroundings
Because it contrasts it's surroundings enough, you can ask the computer in the system to track this blob, so the system starts moving the ATFLIR to keep the black blob in the centre.
The system knows the direction that the ATFLIR is pointing relative to the plane this are expressed as angles
Because the ATFLIR is important and you the pilot have pointed it at something and asked for it to be tracked the system automatically adds a track to your system so you can add it as a target, you can then designate that track as your L+S which can then allow other things to look it, possibly even your RADAR. Without you doing anything
This says nothing about the actual ability of your weapons to hit it.
 

Domzh

Active Member
this is all very clear to me already.
what isnt clear is how the radar knows what you are tracking with the ATFLIR, when the range is not known.

how does the radar "know" that it actually sees what the ATFLIR is tracking.

does it just shoots EM waves in the same direction and if it gets a return, it assumes it must be the same target?
 

jplaza

Member
anyone has a research gate account or science background to make it more likely to get the pdf from the author when requested?

google shows a preview, it seems they looked into some issues with the ATFLIR.
study was published 2006.

https://www.researchgate.net/public..._advanced_forward_looking_infrared_ATFLIR_pod
You can find it in this post along with more documentation:
https://www.metabunk.org/threads/at...cal-documents-and-resources.11801/post-250970
Paper: https://www.metabunk.org/attachments/gerald-uyeno-atflir-paper-uyeno2006-pdf.44938/
 

Leonardo Cuellar

Active Member
this is all very clear to me already.
what isnt clear is how the radar knows what you are tracking with the ATFLIR, when the range is not known.

how does the radar "know" that it actually sees what the ATFLIR is tracking.

does it just shoots EM waves in the same direction and if it gets a return, it assumes it must be the same target?
See this video tutorial. It might help you:

Source: https://www.youtube.com/watch?v=ChsMcv46Rj4
 

jarlrmai

Senior Member
this is all very clear to me already.
what isnt clear is how the radar knows what you are tracking with the ATFLIR, when the range is not known.

how does the radar "know" that it actually sees what the ATFLIR is tracking.

does it just shoots EM waves in the same direction and if it gets a return, it assumes it must be the same target?
The RADAR doesn't have to correlate a track, a track is a just a record in your flight computer that has a position or angle.

If there's an angle only track on a line from the plane, the planes computer can extrapolate all points along that LOS and see if any existing tracks match that LOS, then correlate.

You can also then ask the RADAR to focus in on that point (it can have its emission profile changed to be more narrow and it might pick it up.

But tracks can comes from multiple sources

Your RADAR
The ATFLIR
Someone else's RADAR
Someone else's ATFLIR
 

Domzh

Active Member

Attachments

  • 938F919E-51FA-45EF-A78E-FAC1C5565FFC.jpeg
    938F919E-51FA-45EF-A78E-FAC1C5565FFC.jpeg
    322.4 KB · Views: 67
Last edited:

Mendel

Senior Member.
google shows a preview, it seems they looked into some issues with the ATFLIR.
study was published 2006.

https://www.researchgate.net/public..._advanced_forward_looking_infrared_ATFLIR_pod
The preview is of the page, not of the paper (if it was of the paper, the google link would go to the full text). It's actually in the abstract of another paper in the "Recommended publications" section.

I find that, generally, people who have tried to support their points on Metabunk with Google previews have not done very well. It tends to backfire.
 

Domzh

Active Member
The preview is of the page, not of the paper (if it was of the paper, the google link would go to the full text). It's actually in the abstract of another paper in the "Recommended publications" section.

I find that, generally, people who have tried to support their points on Metabunk with Google previews have not done very well. It tends to backfire.
i dont try to support anything with google previews. hence why i wanted to find the source of it for verification.

thanks for pointing that out though
 

markus

Active Member
I don't know for certain, but I think the main assumption for passive ranging is that the object isn't moving, as something on the ground would be. That's the only way such a measurement could be made. When I modeled GoFast (less precisely than I did Gimbal), there was a pivot point on the sightline where the object could be at rest. At 12,000 feet, I measured 5 knots. No such point exists in the Gimbal case; the slowest speed I could find is 75 knots around 16 NM. And in Gimbal only one angle is slewing significantly; in GoFast both are.

So, radar, maybe. But Lehto called the range number a guess, and radar wouldn't be a guess. A passive trig calculation, based on both azimuth and elevation slew rate, would be.
The generic condition is that the observer's movement needs to be one order higher than the target. That is, if the target is stationary it's enough to move with constant speed, if the target is moving with constant speed in a straight line it's enough to have an acceleration, if the target is accelerating uniformly it's enough to have a jerk, etc. In those conditions you can write down an ansatz for the target trajectory (e.g. something like x = x0 + vx t, y = y0 + vy t) and minimize the sum of squared errors as in an ordinary linear regression. As long as the aforementioned condition applies the equations will be linearly independent and you'll be able to solve for the object position/velocity/acceleration/etc.
 

jarlrmai

Senior Member
The generic condition is that the observer's movement needs to be one order higher than the target. That is, if the target is stationary it's enough to move with constant speed, if the target is moving with constant speed in a straight line it's enough to have an acceleration, if the target is accelerating uniformly it's enough to have a jerk, etc. In those conditions you can write down an ansatz for the target trajectory (e.g. something like x = x0 + vx t, y = y0 + vy t) and minimize the sum of squared errors as in an ordinary linear regression. As long as the aforementioned condition applies the equations will be linearly independent and you'll be able to solve for the object position/velocity/acceleration/etc.
So given the numbers in the Go Fast video from the ATFLIR could Mick for instance, change his simulation to present the Go Fast scenario. Add in extracted numbers for the jet height/speed etc, then add in a "calculated RNG and V/C" from our best attempt at the equations?

This might not be as accurate (or otherwise) as the (presumed by Lehto to be calculated) RNG V/C we see in the the video if that is calculated from more accurate internal numbers not displayed on the screen which instead shows rounded versions.

This would then give us a ballpark on what the internal computer would have been able to calculate to see how it differs from the numbers on the Go Fast overlay.

Then we can just change it back to Gimbal..
 

Mendel

Senior Member.
So given the numbers in the Go Fast video from the ATFLIR could Mick for instance, change his simulation to present the Go Fast scenario. Add in extracted numbers for the jet height/speed etc, then add in a "calculated RNG and V/C" from our best attempt at the equations?
The issue is that a) the observer path is not well known, and b) the GO FAST UAP may be moving.

Compare:
This animation shows the effect of varying the turn rate on the calculated speed. The blue line is the simple two point analysis of the target object path with jet (red line) traveling in a straight line. The orange line is the projected target object path with a variety of turn rates from the jet.

With this model, the speed of the object could be anything from 28 knots to over 100 knots. This speed is relative to the atmosphere at the level of the jet, and so is consistent with a balloon + wind differential from 13,000 to 25,000 feet. Or a self powered bird.
 

jarlrmai

Senior Member
The issue is that a) the observer path is not well known, and b) the GO FAST UAP may be moving.

Compare:
Okay I remember that now, so the question is then is it mathematically possible for the ATFLIR/computer to calculate a range (even as an 'estimate' as suggested by Lehto) and given the data we have, can we "recreate" that estimate?

Micks diagram as above fixes range (r1) to the value in the video and calculates speed of object.

We didn't accurately grab the observer path because at the time, I guess as the range figure was not questioned and so a straight path with a known range was enough to show a slow object, Mick added a curve to the geogebra to show the effect of a bank was to actually slow the object. But for Gimbal we extracted the bank angles and made more of an effort to get the AoA of the F/18 etc so perhaps with that new technique we can get a better observer path for Go Fast and then approach the issue with the goal of showing that what a calculated range using even the rough figures shown on the overlay would be, or at least discover the missing variable which we either don't have and would be required to calculate range.
 

Mendel

Senior Member.
is it mathematically possible for the ATFLIR/computer to calculate a range
The onboard navigation system (inertial/GPS) would have a quite precise idea of how the aircraft is moving, that'd make ranging ground (motionless/slow) targets easy.

On Gimbal, Edward Current used the clouds to pin down the observer flight path, repeating this method using the sea seems more difficult. (This idea may have come up here before.)
 

Domzh

Active Member
The onboard navigation system (inertial/GPS) would have a quite precise idea of how the aircraft is moving, that'd make ranging ground (motionless/slow) targets easy.

On Gimbal, Edward Current used the clouds to pin down the observer flight path, repeating this method using the sea seems more difficult. (This idea may have come up here before.)
shouldnt it be the same principle?

i assume a wave travels in water just like a cloud travels in the wind
 

JMartJr

Senior Member
shouldnt it be the same principle?

i assume a wave travels in water just like a cloud travels in the wind
I think there are going to be some problems there. Cloud features do grow and shrink and "morph" over time, but generally comparatively slowly with features that can persist for some time. Waves are more, pardon the pun, fluid, as the surface of the sea is often shaped by several sets of waves interacting with one another. If you've spent much time bathing at the beach, you may have noticed how every now and then there is a lull in the waves, or more noticeably every now and then there is a bigger wave that can knock you over. This is the result of several sets of waves of different directions/speeds/wavelengths interacting, as described here in an article on how this process can produce huge "rogue waves." (And also the result of "wave sets," according to my reading looking for a source on wave interactions on the ocean, but that's not relevant to this point.)

Constructive interference. Extreme waves often form because swells, while traveling across the ocean, do so at different speeds and directions. As these swells pass through one another, their crests, troughs, and lengths sometimes coincide and reinforce each other. This process can form unusually large, towering waves that quickly disappear. If the swells are travelling in the same direction, these mountainous waves may last for several minutes before subsiding.
https://oceanservice.noaa.gov/facts/roguewaves.html

Trying to track a particular wave crest is possibly going to be a problem, as it crosses or overtakes other waves and so may get bigger or even disappear as the trough of one wave coincides with the crest of another. And that's assuming that the video quality of Go Fast will allow accurate tracking of the (compared to clouds) smaller and less distinct waves.
 

Edward Current

Active Member
The water is moving a lot faster in GoFast than in Gimbal, something in the neighborhood of one field per second, whereas in Gimbal there were only about 9.5 fields of clouds total. That introduces more uncertainty. We're seeing the GoFast surface at an angle, so this further complicates how the fields are counted, whereas the Gimbal clouds are seen flat. The waves are featureless and unsharp compared to clouds which have definite features you can follow. And other problems discussed above.

So, ironically because the Gimbal case was much harder overall, it would be harder to do an accurate model based on the water movement. It could be done, but I don't know how much more solid information that would provide.
 
Top