RR Rated M For Mature
HOME   rrTV-PHOTO   GALLERIES   MY GALLERY   HELP-FAQ
myHOME PM pmRR MEMBERS 407 ONLINE 80 EVENTS SEARCH REGISTER  START HERE
 
1 page6458 viewsPOST REPLY
Heli-Max . Hobby Hut . Heli Wholesaler

.
.
Aerial Photography and Video > Helicopter Autopilot & Camera Control
 
 
JohnC
Veteran
Location: East Yorkshire, England

My Posts This: Topic  Forum		Hi guys,

Some of you may be interested in what I've been developing for the last couple of years. I have built a flight-control system which I have fitted to a 60 size model heli. It maintains the heli at pilot-commanded angles in pitch/roll/yaw and also performs the functions of an engine governer.

Anyway, the system also controls a front camera mount and camera. As the system knows current attitude, it drives servo motors to stabilise the camera. It also (currently) links to a Kodak DC240 digital camera cross an RS232 link to allow the pilot to select zoom + picture shots. There is also a 2.4GHz video downlink.

Does anyone know of a comparable system elsewhere?. I would be interested to see other examples of systems which are more than just an extra receiver + transmitter.


Check out my web site at www.carvec.co.uk to see some pictures. It's been a seriously hard slog, but it finally works!.


John C
09-09-2001 Over year old.
PROFILE   PM   EMAIL   POSTS   BUDDY   IGNORE   HOMEPAGE  
 
 
freestyle
Veteran
Location: Redmond WA USA

My Posts This: Topic  Forum		help me understand...

The stabilized camera mount is very very cool, but I'm not clear about the 'autopilot' features.

The flybar does a good job of stabilizing pitch and roll attitudes, and a modern gyro does a good job of stabilizing yaw. What does your system add to this? Does the Carvec system allow the helicopter to hold a fixed position?

Thanks.
09-25-2001 Over year old.
PROFILE   PM   EMAIL   POSTS   BUDDY   IGNORE   HOMEPAGE  
 
 
JohnC
Veteran
Location: East Yorkshire, England

My Posts This: Topic  Forum		Hi,

The control system holds the helicopter at a demanded attitude. The Tx stick positions are interpreted as a 'demanded angle'. For example, if the heli has drifted off to the right, I hold the stick over to the left and demand (for example) 5 degrees left roll. While I hold the stick at this position, the system keeps the heli at 5 degrees roll. When I release the stick, the heli automatically returns to the upright attitude. If I start to panic (not an unusual occurance), I just release the stick and use the collective to bring it down.

The key to this is actually sensing attitude relative to the ground. Everything else is enabled by this - true camera mount stabilisation and potentially autonomous flight using GPS position information. I am a long way from this at the moment, but it is possible.

I hope this clarifies things a bit!


John C
09-25-2001 Over year old.
PROFILE   PM   EMAIL   POSTS   BUDDY   IGNORE   HOMEPAGE  
 
 
freestyle
Veteran
Location: Redmond WA USA

My Posts This: Topic  Forum		very cool

Yes, thanks for clearing that up. That really does sound like an improvement. :-)

Does uneven terrain confuse the attitude-hold system? Are solid-state gyros good enough (low enough drift) to let you depend on them alone for an attitude reference? Integrating three-axis sensor signals over time should theoretically give you an attitude reference, I just wonder if the drift would add up over time and ruin the reference.
09-26-2001 Over year old.
PROFILE   PM   EMAIL   POSTS   BUDDY   IGNORE   HOMEPAGE  
 
 
JohnC
Veteran
Location: East Yorkshire, England

My Posts This: Topic  Forum		Nate,

The system is entirely based on inertial sensors - so it works the same at 1 foot or 1000 feet. It does use any of that bizarre 'horizon sensing' technique.

The piezo gyros have been a real challenge to get them into a working system. Their long-term performance is terrible!. The software system uses several techniques to compensate for the gyros drift with temperature. The system also senses the local gravity vector, which is used as a long-term reference.

As it uses standard single-rate gyros, I am eagerly awaiting the next generation of solid-state gyroscopes. Hopefully, they won't have quite so many bad habits!.


John C.
09-26-2001 Over year old.
PROFILE   PM   EMAIL   POSTS   BUDDY   IGNORE   HOMEPAGE  
 
 
freestyle
Veteran
Location: Redmond WA USA

My Posts This: Topic  Forum		next gen sensors

I was going to post the URL for this particular gyro I've been looking at, but I just noticed that they make an accelerometer as well...

http://www.spp.co.jp/sssj/moshon-e.html

I'm sure Futaba's sensors have better specs than what you'll find on the web site, but I have a hunch that SPP makes the sensors Futaba uses. I've never seen one of their gyros drift, though, and the sensor shown might still be adequate for cyclic control. Hope you find this interesting...
09-27-2001 Over year old.
PROFILE   PM   EMAIL   POSTS   BUDDY   IGNORE   HOMEPAGE  
 
 
RC-CAM
Senior Heliman
Location: USA

My Posts This: Topic  Forum		Nice work. Keep us posted on your progress.

Interesting enough, on the RC-CAM forum there had been a couple of casual posts about building an artificial horizon using model heli gyros. This has some shared principles with your stabilization system. I had {naively} suggested that their long term stability would cause some grief. So, I can certainly appreciate the difficulty that you mentioned.

In my reasoning, I was concerned with the loss of the initial horizon reference, especially with the piezo gyro's temperature and general performance limitations, which could lead to system drift. In the case of an artificial horizon, I imagined that a "recalibration" would be required every few minutes (a simple matter of flying level and remotely requesting a new zero-reference).

I even went so far as to research how the big boys did it so well with the old mechanical artificial horizons and auto pilots (which had been in use since WW2). From what I found, they never had to worry about drift. It was reassuring to hear that a bulky motor flywheel and crude electronics could do it well.

The good news is that the latest micro machined piezo gyro sensors have the performance that you need. With luck, all you will need is to compensate for the earth's natural rotation, which is 15° per hour.

Again, your project is very impressive. I would like to invite you to post a few words about it on the RC-CAM forum. Just visit http://www.rc-cam.com/ and click on the "Forums" link. It is a great place to publicize any non-commercial video based project.
10-08-2001 Over year old.
PROFILE   PM   EMAIL   POSTS   BUDDY   IGNORE   HOMEPAGE   GALLERY
 
 
JohnC
Veteran
Location: East Yorkshire, England

My Posts This: Topic  Forum		Hello RC-CAM!

Thanks for your post regarding my project.

In my earlier years, I trained as an avionics engineer for the Royal Air Force where we did a lot of work on the old style (spinning mass type)inertial navigation systems. They were incredibly sensitive and accurate for their day - you could see several small drill holes in the wheel where it was carefully balanced in a 'clean' room. They were also mighty expensive!!.

The compensation for earth-rate is actually dependent on the cosine of latitude - when you are at one of the poles, the earth-rate is zero. This is why inertial platforms require to be told current position before aligning. I still find it amazing that they find their own true-north by detecting the direction of earth-rate movement.

Unfortunately, my system is nowhere near such accuracy. The system currently measures angular rotation to about 1/25 deg/sec. I eagerly await the general availability (and with it, reasonable price) of the new technology gyros. Do you mean the vibrating ring type gyros, or is there a development to the piezo technology?.

Thanks for the info on your forum - I will check it out.


Regards,

John C.
10-08-2001 Over year old.
PROFILE   PM   EMAIL   POSTS   BUDDY   IGNORE   HOMEPAGE  
 
 
penguin
Heliman
Location: Ipswich UK

My Posts This: Topic  Forum		I have only had a brief look at your webpage, and your article in RCM&E October issue, but can your system allow inverted flight? I was wondering how well (if it can) cope with doing a roll, because, as I understand it, your left-right roll is governed directly by the position of the stick, how would you turn the stick upside down?
Your work does sound great though, I would love to be able to play around with the electronics like you have.
Keep it up!
Gareth
10-08-2001 Over year old.
PROFILE   PM   EMAIL   POSTS   BUDDY   IGNORE  
 
 
JohnC
Veteran
Location: East Yorkshire, England

My Posts This: Topic  Forum		Hello Gareth,

You are correct in that the system cannot be sent beyond the end-ranges of the stick movements. This is deliberate as the system was not intended for 3-D flying - it was envisaged as controlling a platform which was primarily for something else (eg carrying a camera).

I am however, working on a simplified version which is intended for people who are learning to fly helis and can't/don't want to have to spend a lot of time learning the difficult skills involved. This will be switchable between 'normal' and 'stabilised' modes in flight. The inertial-reference unit works through the full envelope - so, for example, if the heli is upside down and the stabilised mode is switched in then the system will right it again, and the constraints of the stick-position will be back.

Thats the theory anyway, but I get nervous even thinking about flight-testing that one.

Next months RCM&E should be a bit more interesting as it describes the early flight testing phase and should have some nice pics!.


John C.
10-08-2001 Over year old.
PROFILE   PM   EMAIL   POSTS   BUDDY   IGNORE   HOMEPAGE  
 
 
penguin
Heliman
Location: Ipswich UK

My Posts This: Topic  Forum		Sounds good, I hope o start learning sometime soon, your project sounds like it could be a great tool, if only to boost someones confidence that they won't loose their heli.
Your article in RCM&E says you originally got the idea from seeing someone trying to hover, yet they were bobing up and down. Will your project include the ability to sense the altitude therefore you could set the altitude, or just remove the bobing? I assume that it will just remove the bobing as sensing the altitude would be nigh on impossible to do accurately.

When you were testing it, you had it linked to your PC, was that through the Buddy box on the TX? If so, how/where can you gain information on the protocols the Tx's use in the link?
Gareth

EDIT:
After looking at your pages again, it says that the camera angle is changed if it is below an altitude, so I suppose accurately sensing it isn't that bad after all
10-08-2001 Over year old.
PROFILE   PM   EMAIL   POSTS   BUDDY   IGNORE  
 
 
JohnC
Veteran
Location: East Yorkshire, England

My Posts This: Topic  Forum		I think your typo error about 'won't loose their heli' is quite appropriate when we talk about giving it control of the collective pitch!.

The project currently has a sonic-ranger altimeter fitted which works up to about 14 feet max (depending on the surface below). The range sensed here is very accurate and the resolution of the measurement is down to below 1mm. It can do altitude control automatically within this range, but I like to keep the collective manual (so I can bring it down immediately) during development .

Anything higher needs another sensor type (eg barometric) which will not be so accurate.

The Tx buddy-box interface spec was found by using an oscilloscope and some experimentation. It is not difficult, and once I had it I could synthsise my own pulse-train and feed it into the Tx for transmission.
10-10-2001 Over year old.
PROFILE   PM   EMAIL   POSTS   BUDDY   IGNORE   HOMEPAGE  
 
 


Location:

My Posts This: Topic  Forum		Xmiter freq

What is allowed for xmit freq for these Cam Choppers? ( for the US and maby other places)
01-10-2002 Over year old.
PROFILE   PM   EMAIL   POSTS   BUDDY     WV   EDIT
 
 
1 page6458 viewsPOST REPLY
JR-Spektrum . E-flite . Fast Lad Performance

.
.
Aerial Photography and Video > Helicopter Autopilot & Camera Control
 Print TOPIC Advertisers 

Subscribe to This Topic

Saturday, March 20 - 7:21 am - Copyright © 2000 - 2010 runryder.com | email | link to rr | START HERE | NF