Raytheon's Multi-Spectral Targeting Systems

[RobertBillyBob]

There are some great posts on Metabunk explaining how the Raytheon MTS systems work (OSD, FOVs, tracking modes and other settings etc.) but they are scattered through different threads. It's not always easy to search for them when you are having a doubt or a question.
I thought it would be useful to have a dedicated thread to gather existing knowledge and discuss the technical side of the systems (mainly MTS-B) without tying the discussion to any particular case.


I'll start things up with what I found about the mts-b FOVs, it's kind of reversed engineered from the video that where released :

The corners ("┘","┌","┐","└") on the OSD show the borders of the next level of zoom. This way the operator knows if the object will stay in frame when they zoom in.
Knowing the relative FOVs of each zoom level allows us to guess which zoom level we are currently at in redacted videos by looking at the position of those corners.
For example we know this frame uses the medium-narrow FOV because going from medium-narrow to narrow is the only time the FOV shrink this much in one step :

Using the datasheet, PR060 and PR091 I made "rulers" (see attached files) for IR and FMV modes. You can overlay them to different videos to get the current FOV.

Some remarks from studying a few videos :

• IR and FMV modes don't have exactly the same corner position, so the corresponding FOVs are likely slightly different.
• We don't have many FMV videos so we don't have examples for all zoom levels in FMV. My FMV ruler uses some assumptions that way be wrong.
• "Narrow","Medium" and "Wide" share the same corner position (at least in IR). If the zoom level changes in the video you can deduce the current zoom level relative to the others, but if there is no zoom change I'm not sure we can distinguish them.
• If there is no visible corner it likely means you're at max zoom level (ultra narrow + x4 digital zoom)

 

[RobertBillyBob]

I meant the roll of the camera, but it can't be the drone banking because it's filming out the side, so banking wouldn't make that effect.
I guess it's just the operator of the camera rolling it.

I can't replicate it, only on the straight line, as soon as the drone moves to the side I can't figure out how to

I think it's actually banking the view to match the forward horizon. The bank angle seems to match the change in motion of the plane and the parallax motion.

It's hard to say what it shoudl be doing without more reference footage, but this is interesting.

View attachment 90664

Initially the camera seems to be looking right. The actual horizon is lined up withethe artificial horizon.

View attachment 90665

They then swing the camera more to the rightm, the angle remains the same for a while.
View attachment 90666


But as it passed 90° and turns to face backwards, it flips the other way
View attachment 90667

Without de-rotation mechanisms the camera's roll depends on the drone's bank because of the geometric constraints of the gimbal

I created a gimbal simulation in blender and get those results :

No drone bank, the camera pans from front left to back left and stays level :

45° drone bank, the camera rolls when panning :

Using the camera roll as a proxy for the drone banking angle can work if you want the general shape of the trajectory but it's not 1-to-1 unless there is an additional mechanism.