Armadillo Aero hover video

Cool:

http://media.armadilloaerospace.com/2007_05_12/192secondHover.mpg


story:

http://space.newscientist.com/article.ns?id=dn11851&feedId=online-news_rss20

[astrobrian]

Very nice, it was really chewing things up on the ground for a bit there. Great to see things moving along for them

[braddock]

That was impressive.
Still a bit of mild oscillation, unless John was just trying to minimize the heating of the concrete.

[meiza]

Some slosh.

[savuporo]

i wonder if they ( and SpaceX ) are running a low-pass filter analysis or something like that on their servo outputs, so that the control system can actually detect and understand that its oscillating ?
in current video it seems that it dampens out, but its interesting to know whether the control loop coefficents are just tuned well or does the system actually detect and actively counter oscillations.

[PlanetStorm]

savuporo - 16/5/2007  9:36 AM

i wonder if they ( and SpaceX ) are running a low-pass filter analysis or something like that on their servo outputs, so that the control system can actually detect and understand that its oscillating ?
in current video it seems that it dampens out, but its interesting to know whether the control loop coefficents are just tuned well or does the system actually detect and actively counter oscillations.

Looked to me as if most of the oscillation was due to tension waves in the supporting cables. You can clearly see the vehicle having to support what looks like a large amount of cable mass on the upstrokes. Obviously in free flight, this feedback effect won't be there.

[JIS]

savuporo - 16/5/2007  9:36 AM

i wonder if they ( and SpaceX ) are running a low-pass filter analysis or something like that on their servo outputs, so that the control system can actually detect and understand that its oscillating ?

Seeing the last SpaceX test launch I doubt they are doing that.

[Tergenev]

That is an interesting question. I've been an obsessive observer of Armadillo for about two years now, and have read John Carmack's status posts vociferously. My interpretation of what he's doing in this area is based entirely on trying to understand what the control assumptions are, rather than any direct statements that I can recall. But I'm pretty sure that he's just been attempting to tune the control coefficients. I don't think they've been doing any active oscillation suppression.

A hint at this may be seen in this report from Carmack's flight testing report from last August 31st:

'Flight 11: August 31, 2006 hop1

 

Disabled automatic pumping code.  It may have been doing the wrong thing on the last two flights, causing increased drift.

 

Doubled position hold gains relative to orientation hold gains.

 

Perfect 35 second flight!'

 

But more explicitly from his August 8th report:

 "Quad Hop 4

 

Increased propellant load to 20 gallons of fuel, 1000 pounds GLOW.

   

Steady flight, but increasing oscillations eventually caused a tilt shutdown.  Roll thrusters fired in both directions, behaving properly.

 

This type of divergence is usually a simple matter to fix with a gain change, and I was able to replicate the behavior in the simulator by tripling the polar moment of the simulated vehicle, which makes sense comparing the quad to last year's X-Prize Cup vehicle.  Correcting the behavior in the simulator was just a matter of doubling the angular position gain (and leaving the rate gain alone)."

I suspect that the oscillations were part of the data they were trying to gather - i.e., how well does their control system work?

[Tergenev]

I just have to say, a stable 3 minute, 12 second hover by a system created by a volunteer team spending less than $1.5 million and put together in less than 18 months . . . impressive as HELL.  So tell me . . . .why wouldn't a suborbital tourist ride based on this thing be possible within another year or two . . .and costing only a couple thousand bucks a ticket? Strap a BRS aircraft recover system on the side, a seat on top, give me a helmet, and AWAY we go. :-))

[Martin.cz]

Tergenev - 16/5/2007  10:04 AM
 a seat on top, give me a helmet, and AWAY we go. :-))

Work in progress
http://media.armadilloaerospace.com/2006_12_4/seat.jpg

[Crispy]

It's cheap because
a) There's a lot of voluntary labour
b) There's very little safety or reliability engineering - everything is "on-the -fly"
c) There almost zero administration, marketing, sales, insurance etc. costs.

Or at least, that's how it looks to me. My instinct tells me that as soon as you try and make money out of spaceflight, you have to start paying for an awful lot of stuff that isn't a team of dedicated engineers opreating a skunkworks style development programme out of a warehouse.

[jongoff]

Crispy - 16/5/2007  8:21 AM

It's cheap because
a) There's a lot of voluntary labour
b) There's very little safety or reliability engineering - everything is "on-the -fly"
c) There almost zero administration, marketing, sales, insurance etc. costs.

Or at least, that's how it looks to me. My instinct tells me that as soon as you try and make money out of spaceflight, you have to start paying for an awful lot of stuff that isn't a team of dedicated engineers opreating a skunkworks style development programme out of a warehouse.

While I agree that the fact that most of the labor is volunteer, and that they don't yet have most of the functions they would need for a normal business, I do disagree a little with the safety.  They're do actually do some safety and reliability engineering (or they wouldn't have been able to convince AST to let them fly), and if you've been watching them for a while, its obvious that they are getting a lot more reliable.  They've still got a long way to go before they're a normal business, and sure their costs are going to go up as they make that transition.  They also do already pay most of the insurance types that a normal business would need.

But probably not by as much as you would think.  I work for one of their friendly competitors, Masten Space Systems.  We have four paid engineers (and three semi-paid interns this summer), and a marketing/sales VP.  We basically are dealing with all the things you list above.  We haven't yet flown our first prototype, but we're almost there.  And over the existence of our company, we've spent quite a bit less than the ~$2.5-3M to date John has typically quoted.  We haven't gotten as far down the road as they have, but I can see a well-run business doing things for not too much more than what Armadillo is spending.

~Jon

[meiza]

It is cheap per flight because everything is robust and reusable.
It's also a simple one stage vehicle.
It's of course small and low performance at the moment too.

[Crispy]

That's good to hear, jongoff
Keep at it! Prove my hunch wrong!

[Pete at Edwards]

Good to see such videos. Thanks for the link.