US Army MNMS Nanomissle Launch System

[kevin-rf]

One acronym, AEHF

[mike robel]

10 minutes is a long time when in the fight.  However, before we invaded Iraq the first time, we set across the border from them in our attack positions for about 2 weeks.  Plenty of time to put up small satellites to determine/verify the results of ground recon, UAV, and identify enemy reserves outside of our target acquisition range.

If I need info on the enemy 10 minutes out (which works out to about 2.5 km away at normal march rates - more if you are manuvering) a UAV is going to be the platform of choice.

I can see a use for these small sats in the right circumstances, particularly if we (Army units in theatre) can launch them at need, and they downlink to the user in a time-sensitive manner.  I don't see the establishment of big constellations of these, rather they will be launched in theatre to target specific intelligence tasks that cannot be promptly filled by other Army/USAF systems.  I wonder how long you could expect such a small satellite to remain operational and remain in orbit.

I see this targeted at Division and Corps Commander needs, not so much brigade and battalion.  Certainly not company commander level.

[Jim]

The 10 minute cycle has nothing to do with what is happening around the user.

1.  It does not include ground station setup
2.  It assumes  spacecraft is available and is going to have a ground track near the user.
3. The user knows the targets
4.  The ten minutes is from tasking to receipt of data.  This means that the satellite has propulsive attitude control and cell phone type data rates

[XP67_Moonbat]

Heres a fact sheet on the missile:
http://www.smdc.army.mil/FactSheets/MNMS.pdf

[Robotbeat]

Could it be used suborbitally for single-shot surveillance? Just a thought.

[kevin-rf]

Could it be used suborbitally for single-shot surveillance? Just a thought.

Spensive...

[jongoff]

Could it be used suborbitally for single-shot surveillance? Just a thought.

Spensive...

Yeah, for that you'd want something that was reusable.  That was TGV's business case.

~Jon

[jimvela]

{snip}
At a reasonable altitude, that should be able to get you something that has a decent view on a pretty frequent basis.  Nowhere near as good as the AF birds when they're actually overhead, but having 30-72 satellites in the constellation means you're a lot more likely to get intelligence on demand a lot better. 

The bane of the small observing satellite is:
+  Aperture size
+  Pointing ability
+  Data rate (really rate is limited by available xmit power/antenna gain)

Ideally you have all three, though by definition a small satellite cannot have a large aperture.  To make up for the limited space for optics  require extreme pointing capabilities, which generally isn't available in a small sat.

The optics are generally why optical observation birds are the size that they are.

[jongoff]

{snip}
At a reasonable altitude, that should be able to get you something that has a decent view on a pretty frequent basis.  Nowhere near as good as the AF birds when they're actually overhead, but having 30-72 satellites in the constellation means you're a lot more likely to get intelligence on demand a lot better. 

The bane of the small observing satellite is:
+  Aperture size
+  Pointing ability
+  Data rate (really rate is limited by available xmit power/antenna gain)

Ideally you have all three, though by definition a small satellite cannot have a large aperture.  To make up for the limited space for optics  require extreme pointing capabilities, which generally isn't available in a small sat.

The optics are generally why optical observation birds are the size that they are.

So, for something like the KestrelEye, do you think they're being overoptimistic in their spec sheet (I think they claimed 1.5m resolution), or just leaving out details?  You're definitely in a better position to know on things like this than most.

~Jon

[XP67_Moonbat]

Jst slightly off-topic, but a launcher variant of the Pershing missile was looked at back in the day.

http://www.flightglobal.com/pdfarchive/view/1963/1963%20-%200641.html?tracked=1

The Army, theoretically could've had the sort of launch capability back then that they're looking at now with MNMS.

[Robotbeat]

{snip}
At a reasonable altitude, that should be able to get you something that has a decent view on a pretty frequent basis.  Nowhere near as good as the AF birds when they're actually overhead, but having 30-72 satellites in the constellation means you're a lot more likely to get intelligence on demand a lot better. 

The bane of the small observing satellite is:
+  Aperture size
+  Pointing ability
+  Data rate (really rate is limited by available xmit power/antenna gain)

Ideally you have all three, though by definition a small satellite cannot have a large aperture.  To make up for the limited space for optics  require extreme pointing capabilities, which generally isn't available in a small sat.

The optics are generally why optical observation birds are the size that they are.

So, for something like the KestrelEye, do you think they're being overoptimistic in their spec sheet (I think they claimed 1.5m resolution), or just leaving out details?  You're definitely in a better position to know on things like this than most.

~Jon

Jon, it works roughly like this:

resolution=1.22*(altitude)*(wavelength)/(aperture diameter)

They mention 10 inches as roughly the aperture diameter, which is about .25 m, and for a minimum altitude of 200km (2E5m) and a wavelength of about 500nm (5E-7m) , we should get a diffraction limit of about:

1.22*2E5m*5E-7m/.25m=1.22*.1m/.25=.488m

So, 1.5m seems reasonable enough, and because of the low ballistic coefficient of such a low mass but "large" satellite, I would imagine you'd want to be higher than 200km, probably at least 400km. Plus, your wavelength would also be more like 600nm, and combined with imperfect optics, 1.5m seems right on.
see:http://www.google.com/search?q=1.22*500km*(600nm)/(10+inches) gives about 1.44m.

[jimvela]

resolution=1.22*(altitude)*(wavelength)/(aperture diameter)

They mention 10 inches as roughly the aperture diameter, which is about .25 m, and for a minimum altitude of 200km (2E5m) and a wavelength of about 500nm (5E-7m) , we should get a diffraction limit of about:

1.22*2E5m*5E-7m/.25m=1.22*.1m/.25=.488m

Pure angular resolution isn't the only factor that defines system resolution.  Using only angular resolution implies that  your detector is an array of points, which is never the case.

Typical detectors are CCDs, and the size and sensitivity of the individual pixels are a major contributor to the overall system capability.  Most of the detectors used for this type of application have surprisingly large pixels, and surprisingly long integration times.

You also start to be limited by things like geolocation accuracy and repeatability when you start talking about responsive (near real time) tasking.  If  you can image  1Kmx1Km at .4M resolution, but can only geolocate and point with 12Km accuracy, then you could end up shooting useless targets and/or not know exactly where you imaged.

[Robotbeat]

resolution=1.22*(altitude)*(wavelength)/(aperture diameter)

They mention 10 inches as roughly the aperture diameter, which is about .25 m, and for a minimum altitude of 200km (2E5m) and a wavelength of about 500nm (5E-7m) , we should get a diffraction limit of about:

1.22*2E5m*5E-7m/.25m=1.22*.1m/.25=.488m

Pure angular resolution isn't the only factor that defines system resolution.  Using only angular resolution implies that  your detector is an array of points, which is never the case.

Typical detectors are CCDs, and the size and sensitivity of the individual pixels are a major contributor to the overall system capability.  Most of the detectors used for this type of application have surprisingly large pixels, and surprisingly long integration times.

For a day-time, visible image? That is surprising...

You also start to be limited by things like geolocation accuracy and repeatability when you start talking about responsive (near real time) tasking.  If  you can image  1Kmx1Km at .4M resolution, but can only geolocate and point with 12Km accuracy, then you could end up shooting useless targets and/or not know exactly where you imaged.

Absolutely. But you don't need to say that in your marketing material...

[simonbp]

Pure angular resolution isn't the only factor that defines system resolution.  Using only angular resolution implies that  your detector is an array of points, which is never the case.

Typical detectors are CCDs, and the size and sensitivity of the individual pixels are a major contributor to the overall system capability.  Most of the detectors used for this type of application have surprisingly large pixels, and surprisingly long integration times.

You also start to be limited by things like geolocation accuracy and repeatability when you start talking about responsive (near real time) tasking.  If  you can image  1Kmx1Km at .4M resolution, but can only geolocate and point with 12Km accuracy, then you could end up shooting useless targets and/or not know exactly where you imaged.

If you're only looking for one year of operation, you can probably get away with commercial-grade CCD sensors. 1km square at 1.5m resolution is 0.4 megapixels, and at two pixels per resolution unit (pseudo-nyquist sampling), it's 1.7 megapixels, or less than a quarter of the phone in my pocket.

With such a small aperture and low orbit, the real camera issue isn't resolution, but integration time. That will probably drive them to an astronomical sensor with low-noise amplifier. But again, with ~1 year on orbit, it doesn't need to be rad-hard, so an off-the-shelf commercial astro chip should work.

Geolocation is an interesting issue, and one that I'd solve with a combination of cheap MEMS accelerometers (again, cell-phone tech) for pointing and real-time updates on the orbit from Space Command for position. There are probably other ways too, but the trick is remembering that computing is cheap and light, while specialty sensors (e.g. startrackers) are heavy and complex.

Communications is probably through a mesh network, like many ground-based sensor nets. Specifically, you'd have one or two sats imaging at a time, and the rest acting as relays to a ground station or geosync com sat...

[Proponent]

I imagine this is well off into the realm of science fiction at this stage, but maybe you could have cloud of small sensors that could function as an interferometer.  Quite a few challenges but possible in principle.  Then you could get fantastic resolution and the system would be insensitive largely to the loss of a few components.

[Sparky]

I imagine this is well off into the realm of science fiction at this stage, but maybe you could have cloud of small sensors that could function as an interferometer.  Quite a few challenges but possible in principle.  Then you could get fantastic resolution and the system would be insensitive largely to the loss of a few components.

Every technological feat was science fiction until somebody did it.

I'm guessing that the real push for this launch system might be for purposes other than purely coms or recon. Such small nanosats could just as easily be piggybacked onto other more conventional launches. The selling points here seem to be its small size and mobility. I wonder if a launch system and sat are small enough, it might be possible to evade tracking. Or, alternatively, if the launch is detected, mobility might lend plausible deniability to whom the the launch was from.

[Jim]

1.  Geolocation is an interesting issue, and one that I'd solve with a combination of cheap MEMS accelerometers (again, cell-phone tech) for pointing and

2. real-time updates on the orbit from Space Command for position. There are probably other ways too, but the trick is remembering that computing is cheap and light, while specialty sensors (e.g. startrackers) are heavy and complex.

Communications is probably through a mesh network, like many ground-based sensor nets. Specifically, you'd have one or two sats imaging at a time, and the rest acting as relays to a ground station or geosync com sat...

1.  Cheap MEWS would be accurate enough

2.  Not a given.  The orbit could be such that it is without ground station coverage for many orbits

3.geosync com sat for relay would take a lot more power for the transmitter

[Robotbeat]

Star trackers don't necessarily have to be that big... It's possible to fit a star tracker on a nanosat, for sure (though ultimate pointing accuracy is probably limited by aperture size... should be able to get to just a fraction of an arc minute in a very small package, smaller than a human eye). And a Sun or Earth sensor can be very small (though accuracy/precision is generally less than a star tracker).

[Blackstar]

This has the look and feel of a pet project that is going to die out in a couple of years.  It's got a number of things working against it.  None of them individually are killers, but collectively they could be a big obstacle.  Just off the top of my head:

-the Army doesn't build space systems, doesn't really have the expertise, and always faces the risk of the Air Force stomping on their programs and getting them shut down.  Yeah, we're talking about small rockets and small satellites here, but that makes them unusual and in some ways makes them sophisticated, which is difficult for a service with no experience base.

-the goal seems pretty ill-defined.  There's no real existence proof.  What exactly is this going to do?  Why is that necessary?  And is there reason to believe that it can be done better with a space system than with terrestrial systems?  Or by other similar existing space systems?  And that also touches on issues of turf.  Before the US Army starts operating its own reconnaissance satellites, they will have to talk to the guys at the National Reconnaissance Office, who will fight them tooth and nail unless they are convinced that this new approach is a) necessary, b) a good idea, c) won't undercut existing NRO programs, and d) is better done by the Army than the NRO.

-the technology is new and still mostly-theoretical.  There are some big questions that follow from my previous point--what is it going to do, and can it actually do that?  Large constellations of satellites are neat in concept, tough in reality.  It takes a lot of effort to manage the GPS constellation, or the Iridium constellation for that matter.  Put a constellation that big into low Earth orbit and you have to manage it.  Even if the satellites are simpler, there's still a ground station challenge that is probably not on anybody's radar right now.  (That's pretty common, actually--lots of people like to design paper rockets and paper satellites and think that stuff like command and control and tracking and data reduction are simply "details" that will be worked out by somebody else.)


If you want a sense of what is going to happen with this initiative, you should look at the Operationally Responsive Space experience.  ORS was birthed around 2000 or so, with some plans to bring down launch costs significantly to enable small satellites and rapid response.  I believe that this is where the Falcon 1 entered the picture, with an original launch cost of around $5 million apiece.  (Anybody know what the list price is for a Falcon 1 now?)

What is going on now?  Launch costs for small satellites have not dropped at all, and in fact in some cases are higher on a per-pound basis than mid-sized rockets (which ain't cheap).  ORS has slowed and regularized into what looks like a project to build perhaps a payload or two a year, if that.  And the initial efforts have not been that promising.  (One built satellite will probably never fly.  The first satellite launched actually did not get used for a number of months due to a policy dispute that should have been worked out beforehand.)  There's also some indication that ORS is simply out of political favor right now.  It's sponsors have left the Pentagon, and it has not yet returned results that are amazing enough to convince all the new people.

So when somebody comes in excited and with a great new idea, it helps to wait until they have stopped panting and then asked them to answer all of these kinds of questions.

[go2mars]

-the Army doesn't build space systems, doesn't really have the expertise, and always faces the risk of the Air Force stomping on their programs and getting them shut down. 
-the goal seems pretty ill-defined.  What exactly is this going to do?  Why is that necessary?  And is there reason to believe that it can be done better with a space system than with terrestrial systems?  Or by other similar existing space systems?  And that also touches on issues of turf.  Before the US Army starts operating its own reconnaissance satellites, they will have to talk to the guys at the National Reconnaissance Office, who will fight them tooth and nail unless they are convinced that this new approach is a) necessary, b) a good idea, c) won't undercut existing NRO programs, and d) is better done by the Army than the NRO.

If a foe sets off an EMP, or otherwise removes the majority of US satellites from usefulness, you can bet that the army would rather have a bunch little crappy cameras with them in the field, ready for quick-launch than nothing at all.  Plus at 10 kg, they would be smaller targets if it was ground based ASAT.  If the satellites are taken out/blinded by other little enemy satellites (rumoured to exist), then you might be able to have more cameras to launch than they have of little predator satellites.   

It also could be used to test technologies for small hypersonic ground-ground or ground-air missiles (scram/ram).

  I believe that this is where the Falcon 1 entered the picture, with an original launch cost of around $5 million apiece.  (Anybody know what the list price is for a Falcon 1 now?)

Falcon 1 is a lot bigger/less transportable.  It costs $10.9 million for 1010 kg.  This other system is talking about ~10 kg.  One needs liquid oxygen, the other just needs ethane and nitrous oxide.  Far different uses/mobility levels.

My best guess is that its to see how cats like space.