camera stabilization?

[apcopresta]

has anyone tried to build any sort of camera stabilization device for their uavs?

a friend and i were thinking of maybe trying to build some sort of stabilization device w/ a camera mounted in a box w/ a standard RC gyro. the gyro would then drive a servo that controlled the rotation of the box w.r.t. the roll axis of the uav.

anyone tried anything like this? or have any better ideas?

--tim

[kd7ost]

I think the problem with using standard rate gyros, is even though they resist motion from the current axis, they also accumulate error. Once it isn't straight up and down anymore, it will resist motion from it's new location. It won't hold your camera looking straight down. I don't know of any magic formula's. Thats kind of the holy grail in all of this UAV and Camera stuff. Take a look at this product.

http://www.u-nav.com/picotilt.htm

Dan

[radiohound]

This can also be done with thermopile sensors (infrared). A microcontroller would be used to control two servos to keep the outputs of the thermopiles a constant voltage. It is mentioned in section #9 and shown in figure #8 of this paper:

http://www.ctie.monash.edu.au/hargrave/horizon_sensing_autopilot.pdf

[sesat]

If you already are used to panning and tilting a camera platform while in-flight, a good HH gyro would provide great short-term stability. As Dan pointed out though, don't expect it to point in the same direction for long.

Ram.

[gooseuk]

~Hey,

Why not just use one of those Co pilot systems, they work on the princable of keeping the plane level using the horizion, why not just hook your camera on to a wee platform under your plane with two servos and then hook them up to a spare copilot unit.

HEY you just got an auto leveling camera platform

- Phil

[kd7ost]

~Hey,

Why not just use one of those Co pilot systems, they work on the princable of keeping the plane level using the horizion, why not just hook your camera on to a wee platform under your plane with two servos and then hook them up to a spare copilot unit.- Phil

It should work out just fine. We've talked about doing it but your plane has to be structured in a way to support the process. The sensor has to be attached to the camera pod. I keep my camera's inside my plane so it will take some re-structuring to get it to work. I also look straight down so the easiest way to build the unit would be to have the sensor on top of the aircraft. Behind the lens. That means in my case having an attachment from the camera pod sticking up through the top hatch. It would be a better choice in my world to use the unit from U-nav listed above. It can sit inside the plane and be located anywhere on the camera pod as long as I observe orientation. No calibration required etc.

Another point against the co-pilot in this particular case. My camera is fixed mounted and I have no spare channels. I'm already using 8. I use the co-pilot on my airplane for stabilization already. I use a barometric sensing altitude lock on the pitch though and not the co-pilot for pitch. This does a fair, but not perfect job of keeping my camera pointed straight down. To employ a second co-pilot in pitch and roll, you need to have a micro controller or discreet components developing the input pulses to the second co-pilot so it's input requirements are met. It will take two channels of radio (or some other above described process) to do it. You can trick it in some cases buy using your aileron as the roll input to the unit. But, you can't trim it independently that way. Anyhow, it's doable, but by the time you get it all set up that way, that pico tilt is a more affordable solution and a lot more hassle free to install and operate. Smaller parts count, less space, and it's instantly ready to go every time you apply power. That’s just my two cents worth. I don't work for U-nav, and I do use and like FMA's co-pilots. ;-)

Dan

[apcopresta]

cool thanks for the links! the pico-tilt looks pretty cool. i guess the only way to use gyros would be w/ a HH gyro, and would probably need a man in the loop, to periodically re-aim the camera.

--tim

[Duke58]

I like the Co-Pilot platform but my idea is to use a second radio for the platform , also you do not have to put the camera with the sensor head. If you Y off of the servos you can drive a second platform for the camera. I want to put a stabilized camera on the T-tail of a 3M sailplane, the camera view would see the air frame moving in the horizon field.

Earl

[Arp]

For installation on the tail (without spoiling the center of gravity)... it would have to be a *very* light camera (or oppositely a very heavy glider).

I added two little servos directly to the tail surfaces of a standard glider. Oops. Center of gravity *almost* incorrigible (especially since the original NiCad pack in the nose has been replaced by a LiPo).

[Duke58]

Yes , the camera is light and the servos do not need to be in the tail.

Earl

[electroman]

See www.fmadirect.com

Co-Pilot is being used very effectively for both aircraft and platform stabilization. It may be used with a simple pulse generator for the platform if you always want it held perfectly flat or use a second radio and receiver to make it command-controllable. Then the iircraft is stabilzed as usual and can be rolled and pitched at will and the platform stays vetical or at whatever position you command it.

Fred Marks,
FMA Direct