The Little Lofty Looker

Thomas Lord

Fri Nov 07 18:01:26 -0800 2008

manage

It's funny what they leak.

So, you combine those with some (old school) "cloaking" and go to some of the "see through walls" tech and you have quite a little data capture unit. You could put "other things" on the same platform, too.

-t

The Little Lofty Looker

John Dalton

Fri Nov 07 19:27:26 -0800 2008

manage

Maybe wikileaks needs to go on a shopping trip? Current model aircraft electric motors are able to lift the weight of the aircraft or helicopter (allowing a vertical climb). Presumably they could also be used to make a CMOS camera chip with wifi or WiMax hover? The wireless link could be used for command an control, as well as data.

The Little Lofty Looker

Michael Smith

Fri Nov 07 19:59:29 -0800 2008

manage

I have a little electric palm sized helicopter. I don't doubt that a version could be made with better quality batteries and motor, and the payload you suggest for 1000 dollars or so.

In a similar vein Neil Stephenson wrote about micro UAVs in The Diamond Age. They were generally kept afloat by little vacuum flasks. The technology in the novel doesn't allow for Star Wars style force fields but fleets of UAVs perform a similar function by forming a wall which only authenticated people could pass through.

There is a bit of effort going on at the moment to improve aerial dependant survielance surveillance capabilities for aircraft. It is all applicable to unoccupied aircraft and I think we will start to see them over our cities in ten years or so.

State road authorities have large networks of CCTV cameras on freeways. One purpose is to allow early assesment of crash sites so that the right resources are sent in to help. Camera location is never ideal so a movable camera would be a big advantage in this regard.

The Little Lofty Looker

John Dalton

Fri Nov 07 20:09:31 -0800 2008

manage

I presume you are talking about one of those little indoor infrared controlled helicopters, than can be bought for AUD20-50? It would be fascinating (and possibly not even that hard) to equip one with 802.11 and maybe a CMOS camera chip. One would have to cannibalise the camera and wifi, mounting only the bare essentials, to keep the weight down.

If you get a chance, could please you do me a favour sometime and measure what payload your little helicopter can carry and post the result here? You've got me curious now. One method might be to attach paper clips until it cannot lift off, then weigh the paper clips.

The Little Lofty Looker

Michael Smith

Fri Nov 07 22:46:59 -0800 2008

manage

I don't have a good high resolution balance available so I weighed a bundle of 19 paper clips on a kitchen scale. They came to about 45 grams. That comes to 2.4 grams per clip.

With one paper clip it flies better than with none. It is much easier to control, probably because of having more momentum.

With two paper clips it can fly but not maintain altitude outside ground effect. Below 10cm or so it can stay in the air.

With three paper clips it can take off from the ground and fly in ground effect to about 2cm altitude.

With four or five clips on board it behaves like a hovercraft. It can't fly at all.

I think the designers have built these toys to have just power to fly around the house. If it had excess power they would have put in a smaller engine and saved the mass. My guess is that a device this size could be built to carry 50 grams or so of dead mass. There are no COTS components inside these devices. They are highly integrated purpose built toys, which saves on mass.

Given the state of the art it might be better to avoid COTS networking hardware. LEDs are very light weight. I have heard about little camera-on-chip devices for security applications.

This product might be  better model for a UAV, or possibly one built with beefier components.

The Little Lofty Looker

Michael Smith

Fri Nov 07 23:47:24 -0800 2008

manage

Also it occured to me that you could buy four cheap helicopters ($25 each) and strip out the motor, driveline and rotors. Then build a light chasis (balsa wood might do) and attach one to each corner.

You would have to build your own networking and control hardware. But going this way it might be possible to use commercial components.

My original helicopter came from Paddys market in Sydney.

The Little Lofty Looker

John Dalton

Sat Nov 08 04:23:24 -0800 2008

manage

Thanks heaps! Those are interesting numbers. If the battery is custom built, I imagine they made it as big as possible, leaving little excess lift.

Thanks too for the Jaycar pointer. Hopefully I can get one even cheaper through Electus, their wholesale arm.

There are a few options for hardware, some of which you have already covered:

• COTS
• Buy COTS simply as a source of components, then rebuilt it. Who needs things like cases and connectors?
• Custom

An of the shelf 802.11b PCMCIA card weighs 165g. USB might be lighter. I also have an 802.11b transceiver that I built a few years ago, based on a MAX2822 chip. The PCB is 127mm x 23mm x 1.6mm, has quite a number of connectors and weights 65g. It should be possible to build a stripped down version that weighs less. Maybe even a rats nets using bare (packaged) chips? It depends on whether the chips can still be obtained. All food for thought and theoretical until I get my thesis submitted.

alien invaders!

zogger

Sat Nov 08 11:12:21 -0800 2008

manage

Honest, yer honor, I thunked they was dang aliens invading! Looked like a dern UFO thingamabob! Good thing I had me goosegun handy to take 'er out afore they done snatched up some people or mutylated some cows er sumpin'

No, I don't have one..yes I would like one..dang fast moving high flyin alien invaders!

The Little Lofty Looker

duhavid

Sat Nov 08 13:13:04 -0800 2008

manage

"and attach one to each corner"

I would be concerned about stability with that arrangement.

The Little Lofty Looker

Michael Smith

Sat Nov 08 18:24:01 -0800 2008

manage

It is definitely going to need active stabilisation to keep it close to level.

Yaw is an issue too. The UAV will tend to counter rotate. You could reverse half of the rotors but that would require making new hardware so I would suggest aligning an opposite pair of rotors to offset this torque.

That way independent control of thrust from each rotor should be enough for three axis control over attitude.

The Little Lofty Looker

John Dalton

Sat Nov 08 18:57:27 -0800 2008

manage

Perhaps yaw isn't an issue? Don't most of those el-cheapo helicopters use contra-rotating rotors?

I gather active stabilisation isn't that big a deal for helicopters either. All you do is get a small photodiode array (say a central diode, with 4 diodes around it, so they look like a '+' sign). Put a fish eye lens in front of the array and point it upwards. The horizon gets projected onto a circle, with sky forming a nice bright interior to the circle. Stabilisation reduces to keeping the circle centered on the central photodiode. With 4 rotors, one on each corner, a straightforward feedback loop from each of the 4 corner diodes to the corresponding rotor might do the job.

Maybe that can be reversed, pointing the camera down so the horizon forms a bright annulus, which is then to be centered? That way the one camera can perform stabilisation while also imaging the earth below.

The Little Lofty Looker

President4242

Mon Nov 10 08:32:06 -0800 2008

manage

I think this flying saucer or this blimp would be better choices for homebuilt video UAV.  The flying saucer's autotrim feature, and the manual installation of helium in the blimp, could be used to adjust for weight of the camera.

Parking

tallfred

Mon Nov 10 08:38:09 -0800 2008

manage

You need to put a big suction cup on the side, so it save gas by holding on to a bridge or building window using a vacuum system.

Parking

John Dalton

Mon Nov 10 13:55:02 -0800 2008

manage

Yep! I proposed an idea a little like this once on the "Sydney Wireless" mailing lists. The problem was how to deploy a free wireless network in the absence of sites to locate access points.

The suggested solution (never implemented) was to cross a self contained access point with a limpet mine. Access points were to be solar powered with a strong magnet attached. It would then be slung from a model helicopter and flown close to a nice high metallic object, to which the access point would stick then release. Chances are the owner of the tall object would never even realise the access point was there. Even if they did see it, it would probably not be worth their while to remove it.