controlling UAVs/RPVs at 900Mhz?

[dalbert02]

I was thinking it would be nice to get away from 50Mhz/72Mhz to control my UAV/RPV. I don't want to go to 2.4Ghz since much of my video equipment is 2.4Ghz. I was pondering the idea of using maxstream's serial radios that operate at 900Mhz and using a program controller like this:
http://www.servocity.com/html/servo_controllers.html
It would be nice to plug in a R/C controller with USB into a laptop, then take the RS-232 serial out of the laptop to drive the servos. Obviously some programming would be needed, but what do you think?
-dave

[Myron]

Dave,

What kind of system are you using now? We use a Maxtreme Xtend and ours can be flown that way on an off the shelf USB joystick. Of course all of this is user configurable with the AP software. The system can be flown via the mouse as well but if there is ever a problem and you want to take manual control its easiest to do it with a joystic.. .We use a couple different ones from Saitek and microsoft..

Myron

Myron

[dalbert02]

Right now I am using a 9CAP but I would like to get away from 50/72Mhz. I see there are USB servo controllers as well, so I could go USB all the way and get away from RS-232 completely. I guess what I need to do is figure out how to take the USB out of my R/C joystick and port that to the USB out of the PC in such a way to control the aircraft. Thanks,
-dave

Something like this:

TX to USB

PC USB to servo driver

http://www.obdev.at/products/avrusb/projects.html

[danstrider]

Cloud Cap's Piccolo already is on 900MHz. You're in good company!

Dan

[dalbert02]

Well at $5500 to $7500 they may be a little out of my price range at this point. I need something a wee bit cheaper. Thanks tho.
-dave

[dalbert02]

I think I may have found a solution:
http://www.ringolake.com/pic_proj/servo/rc_trans/rc_trans.html
This device with a pair of RS-232 radios should work!
-dave

[1stCAVgrunt]

Ok, I've got planes, and I've got a sweet laptop to use for a ground control station.....what other than a radio modem and what I've already got, do you need to make your aircraft semi-autonomous? What kind of low end software and hardware do you need to use to get it off the ground with your computer being your transmitter? I'd like a system where I can use a joystick so that I can also control cameras on the aircraft while flying. Thanks

[danstrider]

Aw man, just had the thought this would be an AWESOME way to do autonomous vision-based piloting offboard .... so send a cockpit view live video feed from the aircraft to a receiver on the ground that is hooked up to a computer. The computer does image processing magic based on the image, and then can send control stick information via serial back up to the aircraft! So it's not like that will be easy, but it sure would be a fun way to have oooooodles of real-time processing power and yet still do image-based piloting. It would be like NASA's Autonomous Air Refueling Demonstrator, except that all the video processing would be done offboard and there would be absolutely ZERO sensor data :-)
http://www.dfrc.nasa.gov/gallery/Photo/AARD/index.html

Neat idea for serial control of the aircraft even without the image processing stuff. I'm going to keep my eye on this one.
Dan

[1stCAVgrunt]

That's kind of what I'm workin on. I'm in Iraq right now, and UAV coverage sucks...so I would like to use my experience with RC planes and couple it with my laptop to get greater range out of my planes with the capability to do video recon of our AO. I've already played with the 2.4 ghz cams on my planes but it's kinda crappy to be sittin there with your Tx and watching a grainy tv. Plus if I could use joystick and thottle and even game pads, I could train up my guys to fly it as well. Heck, if half of our equipment was run with game pads, all of our nintendo generation soldiers would be better at what they do ;)

[workshop]

Dave,

I've been thinking of this for gimbal control when the UAV is beyond R/C range (since R/C is just used for TO and landing). I'd be interested in any specific links you find. I wonder if one can get GPS down at the same time one is sending servo serial up since the Maxstream modems are bidirectional?

Myron, Is there a link to your Maxtream setup?

Thanks,

Jeff

[workshop]

Whatever happened to this thread?...

I have a servo controller (Parallax I think) and a Basic Stamp. I’m looking for a low cost semi-industrial joystick that I can use. I prefer not to hack a RC TX and a quick search revealed some pretty expensive joysticks! Anyone know of a good stand alone joystick with pots?

Jeff

[camship]

See if any of these fit the bill.

Here is the makers page: http://www.ctscorp.com/components/pointing_devices.asp

Get the "252" models here:
http://www.digikey.com/scripts/DkSearch/dksus.dll?Cat=262970;keywords=joystick

$5.75 - $ 6.00 each
Good luck

Camship

[typicalaimster]

Whatever happened to this thread?...


Donno, but if he's still using this controller (http://www.ringolake.com/pic_proj/servo/rc_trans/rc_trans.html) he might want to omit a few lines of code in it. Specifically the ones dealing with 'compression'. You'll want to send up everything instead of just the channels that have changed. I ran into an issue with certain channels bleeding over but not changing back.. For example when I'd move channel 3 it'd somehow end up moving channel 1. Probably something to do with my TX and usually not a problem in PPM. However, channel 1 wouldn't be resampled until I moved that stick. Thus I ended up with an aileron that was drooping. Bad Mojo when you go to fly.

I've since removed the lines of code and retested on the ground. Everything so far looks good for another aerial test. I just haven't had the time to pick up a new plane.

[vector_vortex]

My major project at uni this year is exactly this kind of setup.
I have a board using an Aerocomm 1W 900Mhz AC4790 radio modem, With GPS and pressure sensor built in, It has two servo outputs but that could easily be extended to more. I have the output lines in place already.
I have a gimballed directional antenna on the ground which is automatically directed based on the GPS and pressure differential between the ground and in-flight systems. My system also implements a mesh network so you can control up to 254 UAV's.
I also have some rudimentary ground control software done in java.
I expect to be finished coding by end of september and the project needs to be complete by oct 23rd. I am currently developing and testing antennas. I'll be looking to release my final report to anyone who wants to look at it.
If you need any info on what I've done so far just ask.
Have Fun.
Joel

[workshop]

Joystick... Thanks Cam, but I was thinking more of a complete unit:

http://www.etisystems.com/j34.asp

I fly a preprogrammed UAV beyond RC range and I hate loosing my gimbal (it fail safes to a forward down looking view). Since my Xtend is sending down 4800 baud GPS, I figured I could send up 4800 baud gimbal moves (two channels) to operate the camera at all phases of the flight plan. I'm going to start the BasicStamp code this weekend and hopefully port it to an ATMEL chip in the future.

Jeff

[typicalaimster]

Have you tested the Maxstreams with full duplex operation? I noticed you can only send one way with the Xbee's and Xtreams. I could send up stick movements via serial, but my GPS downlink was queued. The inverse would happen if the GPS was transmitting on the wire when I'd send up stick movements. The overall result was control lockout while the GPS was transmitting down. Then a stream of very lagged stick commands transmitted up once it was done.

[workshop]

typicalaimster,

Now that you mention it, I seem to recall the Maxstream rep telling me the same thing; that the modules don't use a "true" full duplex scheme. :rolleyes:

So I imagine a handshaking routine that keys off of the GPS data stream would be one possible solution. The ground station would wait for a pause in the 1Hz GPS string and then upload joystick positional data during the pause. Hopefully the joystick data is small enough to transmit before the next GPS string starts coming down the wire. The servo movements can then be dampened by software on the airborne receiver/servo driver by queuing up position data.

Given the ify nature of this I may get one of those little joysticks that Camship pointed out. The nice ones are over $100... too much for a breadboard... Thanks Cam.

Jeff

[AnthonyRC]

This might be relevant to this thread.
http://www.rcgroups.com/forums/showthread.php?t=728111

I've just completed the PCB design for a 10 channel receiver based upon the AeroComm modems. The prototype systems have worked fantastically, with low latency control, at 5km+.

[n0qbh]

Donno, but if he's still using this controller (http://www.ringolake.com/pic_proj/servo/rc_trans/rc_trans.html) he might want to omit a few lines of code in it. Specifically the ones dealing with 'compression'. You'll want to send up everything instead of just the channels that have changed. I ran into an issue with certain channels bleeding over but not changing back.. For example when I'd move channel 3 it'd somehow end up moving channel 1. Probably something to do with my TX and usually not a problem in PPM. However, channel 1 wouldn't be resampled until I moved that stick. Thus I ended up with an aileron that was drooping. Bad Mojo when you go to fly.

I've since removed the lines of code and retested on the ground. Everything so far looks good for another aerial test. I just haven't had the time to pick up a new plane.

Your issue with servo "droop" is something we never considered when developing the software.
Maybe if there was a mandatory function in the compression routine that would insure all channels get updated at a predetermined interval...
What data rate are you transmitting at?
The compression was added to lower the response lag in the first 4 channels when we were using 9600 baud Max Stream radios.
I'm also curious about the channels mixing up.
I don't recall ever seeing a problem like that. I'll have to wipe the dust off of that file and see if I spot a potential bug.
If I decide to do a rewrite of the code, what else would you like added?
Mike

[typicalaimster]

I'm actually just transmitting at the same 9600 baud over the Maxstreams. If I was mixing 4->5 I wouldn't have a problem.. However if I was mixing 4->5 and 4->7 I'd start seeing some erratic behavior between the other channels. For example with 4,5,7 mixing, channel 3 would behave sporadically.

From what I can tell the circuit works just fine. It's just the little quirks.

[n0qbh]

I'm actually just transmitting at the same 9600 baud over the Maxstreams. If I was mixing 4->5 I wouldn't have a problem.. However if I was mixing 4->5 and 4->7 I'd start seeing some erratic behavior between the other channels. For example with 4,5,7 mixing, channel 3 would behave sporadically.

From what I can tell the circuit works just fine. It's just the little quirks.

Could you explain what you mean by mixing and what it's purpose is?
I'm just getting into RC myself - in the past I was just building circuits to order.

Mike

[Unterhausen]

mixing is a transmitter function that adds two control inputs together. As an example, let's say your aircraft pitches up too much at high throttle, and you're too lazy to fix the thrust line. You could take the throttle input and mix some of it in with your elevator so high throttle gets some down elevator. It's called a mix, because the elevator control is still active, the elevator command the Tx sends out is the sum of the two inputs.

[typicalaimster]

Or in my case I have a Futaba 7c and use it to fly FPV. I have my camera pan tied into my nose gear stearing. I use channel mixing between channel 4 (Rudder) and 5 (Nose gear stearing) to pan my camera. This is assigned to a toggle switch so I can turn it on or off. Then I have another mix for rudder. This is from my rudder stick (Channel 4) to channel 7. I've also assigned this to a toggle switch.

During flights I'll turn mixing on and off depending on what I have going on. During takeoff I'll mix 4,5, and 7 together. After I'm airborne I'll switch off my rudder and just control the camera.