11 replies to this topic

[mike_s]

http://myauntishot.c...and-helicopters

So I had my basic idea down. iPhone joins the Linksys router network. It gets an IP address. Then, I open up my pilot program. The pilot program interfaces with the router via SSH (I couldn’t think of a better way that has redundancy, and speed, and was already buily by someone else). The pilot program interprets what the iphone is doing, and outputs data to one of the ethernet ports of which there are conveniently 4. Rudder, Ailerons, Throttle, Elevator.

Finally, you may ask how the servos are being driven. Well, routers are used to send bits of information down a series of twisted pair wires usually. Guess what it takes to send packeted information? An IC that would work really well as a PWM! I did some haxoring around on this, and read what other nerds had done on the internet, and the next think you know I have a servo with a Cat5E plug on the end of it.



There seems to be some "speculation" on whether this is actually doable

http://www.helifreak...759#post1287759
http://www.rcgroups....d.php?t=1043150
http://runryder.com/...pter/t512044p1/


I'm curious (along with others) on the ethernet -> PWM -> servo interfacing.

What is required for the hardware / software to control multiple servo channels over ethernet?

The pilot program interprets what the iphone is doing, and outputs data to one of the ethernet ports of which there are conveniently 4. Rudder, Ailerons, Throttle, Elevator.

I don't see how seperate ports on the router is of any use, unless there is a seperate i.p. address / network card attached to each servo ?

Mike

[Kilrah]

Possible - yes of course. Now as to whether this particular occurrence is real or not, I'm 99% sure it's fake, so I doubt he'll respond to your comment on his page

The guy is clever as the "research" around the thing is mostly coherent, but some details are definitely off.

- As much as the (introductive) research is detailed, the practically pertinent development (what he would actually have done to make it work) has been mysteriously "forgotten". When asked about details, he just gives very vague info, linking to generic websites that could be more or less related, but nothing specific.
- The "bench test" shows variable latency. To me, the guy is moving the stick with his other (hidden) hand. We don't see hardware on the heli either (a router is fairly big, with 2 antennas, etc, hard to hide on a 450 heli).
- On the flight test footage, the movements of the guy don't match those of the heli, really looks like he's watching the heli that is flown by someone else out of camera view and making exaggerated tilting body movements in response to the heli's movements. He's always a bit late, and doesn't even seem to realise when the heli has landed. No hardware sticking out either.
- He shows a throttle that would be controlled on the iphone's screen. No way he could fly a heli like this, at least that precisely, and wouldn't be holding his iphone that way to use it. Imagine holding it, doing precise tilting movements, while keeping permanent (and just as precise) control on the throttle by sliding your fingers on the screen. Definitely not.
- The heli supposedly is flown using the iphone's accelerometer, which means it will react to accelerations - when during the flight the guy sometimes moves abruptly, the accelerometer would sense that in just the same way as it would sense tilt, and the heli would jump.

I don't see how seperate ports on the router is of any use, unless there is a seperate i.p. address / network card attached to each servo

He supposes that the ethernet ports would be "misused" to each directly output a servo PWM signal, instead of connecting to another ethernet device. I don't know ethernet physical layer chips well enough to say whether this is possible or not, but if it is it would involve extensive hacking of the router firmware. And there would be easier ways too if he really used a Linksys router as he says (there are GPIO ports on the main chip), so I'd say it's unlikely he's do that.

To do this, I'd first work out a way of controlling pitch/throttle that would actually be usable. Then do a little board on the heli with a dsPIC and an ethernet PHY chip. Create a server software that the iphone could connect to that would take network data and generate the servo signals. To retain the linksys AP idea, that board would be connected to one of the router's ethernet ports.

[Mr.RC-Cam]

In the video he said he was a competition R/C pilot. I did a bit of Google'n to see if there was any mention of his name in conjunction with R/C contests. Nothing but an unrelated video of a dead stick landing and miles of hits about the iPhone app.

But, I found his original post on rcgroups. He has a witness to the iPhone flight that posted there too. As with the other linked discussions, there was some comments posted about us being punked. I'm on the fence; It's all possible, but I'm not sure that what we see is 100% legit. Could be, but I also can appreciate a good joke and this one might be a late April-1 inspired attempt that has gone viral.

http://www.rcgroups....d.php?t=1042266

Josh appears to be located in Chandler, AZ. Wouldn't take much for a trusted R/C'er in the area to invite himself to a demo. But according to the friend/witness that test flew it, he is drunk and/or hungover on the weekends. So, schedule your visit for a weekday flight.

[christinedwrds]

looks like a good idea to me. But I don't really know if its doable. If it was real, we probably are testing it already. I just hope someone can truly get this working!

[zeldorf]

Well, that guy may be faking it, but I just saw this on Slashdot and the Iphone bit reminded me of this thread.

http://www.parrot.co...rot-ar-drone/en

The Parrot website seems to be under a bit of strain at the moment, here's the news article I saw it in, with videos, http://recombu.com/n...ess_M11288.html

Looks pretty neat!

[Kilrah]

To me these are faking it just as much... look at what the iphone is doing and the quadro's flight. Just nothing in common.

[Mr.RC-Cam]

Yes, the pilot's movements seem to be a bit odd. But, it was a product advertisement video and nothing is real in those.

As it turns out, they will have the drone at the CES 2010 event, which opens tomorrow. I'll be there and they are one of my vendor stops. If the line is not too long, I will try to get some hands-on FPV time with it.

[Mr.RC-Cam]

I flew the Parrot drone today and it indeed is everything you see in the demo video. Their booth was always packed and a huge buzz at the event. Every dude stopped to gawk at it. And their must be 100,000 dudes roaming these hallways.

The iPhone interface is easy to use. The model is essentially autonomous and the remote control is secondary. There is a control engage soft button on the iPhone that you depress during remote operation, which then allows the model to operate in a fly-by-wire mode. Pitching and tilting the iPhone does indeed allow for precise control. Yaw and altitude are via on-screen soft buttons.

The real-time video looks like about 5 frames/sec (best guess), so that feature does not earn any bragging rights. Range is a bit short, perhaps 25 meters. It also works with the Apple iTouch media player, so you aren't stuck using an iPhone for the remote control mode.

{Edit/Update: I later learned that the slow video frame rate was due to streaming the video from the iPhone to the wall mounted monitor. If the streaming is not used then the iPhone's or iTouch's displayed frame rate will be 15-30 fps.}

The France based vendor was very hush-hush about release date and price. But I learned that it will definitely come out this year (no formal date was given) and will be marketed as a virtually augmented game. Hardware will remain proprietary, but a gaming SDK will be available to developers. Too bad they don't plan to release the hardware/firmware info, but I suspect that hackers will unlock the details to it.

Despite the secrecy, I saw that it uses a 3S/1000mAH pack, which gives 10-15 min flights. I also noticed a couple ultrasonic transducers on the bottom. I'm not sure why they are needed since I heard it uses a vertical mounted camera for altitude references.

I have some video I took. I'll get it posted when I get back from the show. For now, I'm about to crash after a day of looking at really cool consumer electronic gadgets that have been secrete until this week.

Edited by Mr.RC-Cam, 10 January 2010 - 03:43 PM.
Frame Rate clarification

[W3FJW-Ron]

Did ya manage to catch the 3D TVs? Maybe some day they'll be available in FPV goggles. 3D FPV combat.. YEAH!!

[Mr.RC-Cam]

3D TV was the big new technology on display. Works excellent, at least as good as I saw a few weeks ago at the local movie theater showing Avatar in 3D Digital.

Passive glasses were the main viewing method. One vendor I spoke with said the lenses in the glasses are circular polarized. I believe electronic shutter systems are being demo'd too, but I did not see any today.

[Kilrah]

Did ya manage to catch the 3D TVs? Maybe some day they'll be available in FPV goggles. 3D FPV combat.. YEAH!!

There have been 3D googles for a long time, and it's particularly easy as you already have a screen for each eye anyway. Simply need to display something different on each.
Some of the popular FPV goggles support it.

[Mr.RC-Cam]

I've created a dedicated Parrot AR.Drone topic, so that is where I will post my videos.
http://www.rc-cam.co...__0