Autonomous parahutes hit the mark !

[cobaltdan 0]

air dropped from 12,000' & 2 miles away Atair's autonomous guided parachutes are becoming increasingly accurate. Recent drops hit 9 meters and 27 meters from the target !!! These systems fly fully autonomous and are not remote controlled.

upcoming tests will feature a fleet of systems in the same airspace, communicating under an rf p2p network and performing collision avoidance using swarm algorithms.
sincerely,

dan preston <><>
www.atairaerospace.com

Daniel Preston <><>
atairaerodynamics.com (sport)
atairaerospace.com (military)

[PhreeZone 20]

Would it not be easier to deploy and get better proformance if it was freebagged? That huge drogue has to eat into the proformance on that canopy.

Yesterday is history
And tomorrow is a mystery

Parachutemanuals.com

[piisfish 140]

what is the load suspended under the rig ?? How heavy ??

Do you plan to sell the system to skydivers who always land off the DZ ??

scissors beat paper, paper beat rock, rock beat wingsuit - KarlM

[mr2mk1g 10]

My take on it is that the "huge drogue" is there to overcome the difficulties of autonomous flaring. IE the technology that would judge a flare height accurately enough to effectively protect the cargo is not yet economically available – a way round this would be to simply design a system that turns your controllable ram-air into a conventional round for landing.

So I take it that round simply deploys at 50ft or so and lands the thing?

[PhreeZone 20]

In this picture http://www.dropzone.com/cgi-bin/forum/gforum.cgi?do=post_attachment;postatt_id=26641; You can see the drogue trailing off the center of the canopy. The way the Onyx works is you toss the bundle out from any altitude and it freefalls under a tandem drogue until it opens the canopy, then it flies around under the square to a certian altitude that makes the canopy go into a rapid diving spiral. In this spiral the round canopy (basically a Paraglider reserve from the looks of it) pops out and the square it taken out of the picture. The entire bundle lands under the round.

In the picture you can see the drag that the towing drogue is producing. The outer end cells are going faster then the center cells to form that bow shape, or else thats the only reverse cut ellipse Cobalt I've ever seen. Since the center has all the drag its being slowed and the outer cells remain at normal flight speeds until they slow down to the speed of the center cell again. You get to see the effects of a non collapsed drogue on tandems do the same thing.

Mainly off the shelf parts to make up the parachute portion or nearly the same designs. Harness is different then a skydigin one, but probally same thing, 2 containers, spring loaded reserve, etc. Not a bad idea at all.

Yesterday is history
And tomorrow is a mystery

Parachutemanuals.com

[mr2mk1g 10]

I see what you're referring to now.

[pilotdave 0]

Cool but definitely not a new idea. Here's a quote from a 1968 technical paper on the development of ram air canopies (paper available HERE):

"The present breadboard automatic guidance and control subsystem for the 500-lb payload system operates such that upon sensing that it is of[f] course by a predetermined amount, the proper steering line is pulled full and is not released until the unit senses that it is again back on course. This results in the system flying somewhat of a sine wave flight path towards the transmitter. To compensate for the relatively slow speed of the available servo motors and this overcontrol dictated byt he available guidance electronics and automatic homing mode turn rate of 13 to 15 deg/sec was selected and has proven successful in winds of up to 50 fps.

When the 500-lb payload system flys over the transmitter, it assumes an orbit pattern generally resembling a figure eight. Impact accuracies of 200 yd or less are obtained depending on where in this figure eight pattern the unit is when it runs out of altitude. A more realistic impact accuracy, represtantive of that attainable with a guidance and control subsystem properly designated for this application, is that obtained with manual control inputs. With this present hardware, an operator can insert manual override commands as the system nears the ground and can easily increase impact accuracy to less than 50 yd."

So they figured with pre 1968 technology they could be accurate to 50 yards. Almost 40 years later you're still only good to 27 meters??

I really just find it very cool that this technology is back in development. It was one of the original proposed uses for ram air canopies when they were first invented.

Dave

[diablopilot 2]

Quote

Would it not be easier to deploy and get better proformance if it was freebagged? That huge drogue has to eat into the proformance on that canopy.

I thing the drouge gives much more in the way of options for insertion. If you droped the package from 30,000 or 40,000 agl you might want it to freefall a bit before deployment to avoid detection.

----------------------------------------------
You're not as good as you think you are. Seriously.

[FrogNog 1]

Quote

So they figured with pre 1968 technology they could be accurate to 50 yards. Almost 40 years later you're still only good to 27 meters??

The quote you have says that they were getting 200 yards accuracy automously and 50 yards manually. These autonomous 'hutes [sic] are going better than that on their own.

I want to see these bad-boys fly cross country.

Ooh, or CReW!

-=-=-=-=-
Pull.

[pilotdave 0]

Yeah but it says with a better control system (with available technology) they'd be getting closer to 50 yards. But I was kidding... They needed a ground based transmitter in the first place.

Dave

[darkwing 5]

I think the 1968 version required a beacon to fly to. The current ones are gps I bet. No beacon needed.

-- Jeff
My Skydiving History

[quade 4]

So, did the camera flyer dock on it?

On a more serious note. How do you deal with the concept of canopy collisions in a mass airdrop? I would imagine that it would not be horribly difficult to have some sort of low power computer network (801.11b?) set up so that all of the units know where each of the others are, but my guess is that this hasn't come into play yet.

quade -
The World's Most Boring Skydiver

[cobaltdan 0]

the system is modular and can be rigged for payloads from 75 pounds to 2200 pounds. the set up in the pictures is for 75-500 pounds. the drogue shown is 25" in diameter and collapsable. the vector velocity of the cobalt wing is about 80mph and the glide ratio just over 4. with the reserve (round) put out touch down is 22fps.

we build main canopies for extremely high wingloadings. tests have been in the range of 4 - 22 pounds per foot ! the goal is to get from altitude to just above target as fast as possible, then transition to a secondary for a soft touchdown. using a single wing for guidance and landing gives you poor performance for both requirements.

btw, there are actually no off the shelf components what so ever. every component from the actuators to the inertial sensors to the parachutes were designed and built for this project. there were a few stock components in early development but no longer.

i will post more picks when available.

sincerely,

dan<><>

Daniel Preston <><>
atairaerodynamics.com (sport)
atairaerospace.com (military)

[cobaltdan 0]

we developed a custom rf peer to peer network comunications protocol. it is masterless and supports about 50 uav's . 'fleet' tests will begin late this month.

-dan<><>

Daniel Preston <><>
atairaerodynamics.com (sport)
atairaerospace.com (military)

[PhreeZone 20]

The drogue is fine for the freefall portion of the insertion but why have it remain attached to the main after deployment? Whay purpose does it serve to have a collaseable drogue remain attached to the main when you are wanting a gliding canopy?

I know RWS is involved in this project so I imagine the 3 rings are in use like on a Sigma. At release of the drogue to start the canopy deployment make the drogue, bag, etc all turn into a freebag deployment system to lose those drag components.

Yesterday is history
And tomorrow is a mystery

Parachutemanuals.com

[diablopilot 2]

Quote

The drogue is fine for the freefall portion of the insertion but why have it remain attached to the main after deployment? Whay purpose does it serve to have a collaseable drogue remain attached to the main when you are wanting a gliding canopy?

I know RWS is involved in this project so I imagine the 3 rings are in use like on a Sigma. At release of the drogue to start the canopy deployment make the drogue, bag, etc all turn into a freebag deployment system to lose those drag components.

Agreed. For a production model it wouldn't be hard to design the drogue to release after extracting the canopy (make it a freebag) or have the drouge static line a smaller P/C with breakcord. Could get expensive in the testing phase.

----------------------------------------------
You're not as good as you think you are. Seriously.

[PhreeZone 20]

>Could get expensive in the testing phase.


Thats what the camera man is there for

Yesterday is history
And tomorrow is a mystery

Parachutemanuals.com

[billvon 3,090]

> In this spiral the round canopy (basically a Paraglider reserve from
> the looks of it) pops out and the square it taken out of the picture.
> The entire bundle lands under the round.

I would think it would be more reliable to simply cause a B-line stall or something and sink the original canopy onto the target.

[ChasingBlueSky 0]

However, in a mass drop scenario a flock of drifting freebags could cause massive collision issues. Also, if you were going for silent insertion, the freebag system would most likely land first and away from the LZ which could draw unwanted attention.

Good idea as far as flight and drag Phree, but from my civilian mindset it doesn't seem like it would be a good idea.

_________________________________________
you can burn the land and boil the sea, but you can't take the sky from me....
I WILL fly again.....

[Cloggy 1]

I did a few B-line stalls on paragliders: bit scary at first but fun! The descent was about 30ft/s. The reopening is not the best for the canopy, but that wouldn't be important in this application. Maybe B-line stall is not an option because of the higher wing loads?

I'm a bit jealous of Dan's job, sounds like a lot of fun Must be an amazing sight to see a swarm of these things flock together and head to the DZ.

But... if I were a tandemmaster I'd start looking for another job...

[bor 0]

Quote

4 - 22 pounds per foot ! the goal is to get from altitude to just above target as fast as possible,

Dan, what is the glide ratio at these loadings?

[cobaltdan 0]

on the smaller systems the difference in glide ratio with or without the collapsible drogue is very marginal. as a general rule it is best to have all components land together so as not to create a hazard for aircraft .

glide ratios vary depending on wingloading and drag of the load. but with reasonable drag cargos at 5# wing laoding, the systems range from 4 to 6.2 :1

<><>

Daniel Preston <><>
atairaerodynamics.com (sport)
atairaerospace.com (military)