Welcome to our community

Be a part of something great, join today!

Log in
  • Hey all, just changed over the backend after 15 years I figured time to give it a bit of an update, its probably gonna be a bit weird for most of you and i am sure there is a few bugs to work out but it should kinda work the same as before... hopefully :)

High quality but economical gyrostabilized gimbal for Epic/Scarlet

Thanks for everybody for the useful feedback.

Cliff, I'll make a search to find out what has been published on detecting gravitation and stabilization.

The cost of a good IMU system is around 500€, and about the same for GPS modules. What could be useful is the "accumulated phase data" from GPS that yields velocities once in a second with rather high accuracy. In my understanding the accuracy is within couple cm/s, and this is the most accurate data one can get from a GPS. However, accelerations affect the oscillators and make GPSs less accurate. Another issue is, if the stabilization system relied on GPS and one looses the GPS signal during a shot, this easily drives the camera group crazy/angry.

About Coriollis forces, have you seen these big pendulums in any science museum that eventually hit say some bowls? What happens is that the pendulum keeps its track but the earth rotates underneath. Still, don't believe Coriollis forces created any serious difficulties in the end of the day.
 
Another issue is, if the stabilization system relied on GPS and one looses the GPS signal during a shot, this easily drives the camera group crazy/angry.
Click to expand...

Receiver Autonomous Integrity Monitoring (RAIM) provides integrity monitoring of GPS. This function operates when a minimum of five satellites with acceptable geometry are visible to the receiver. Under normal conditions, at most geographic locations on the planet, seven to twelve satellites are in view.

More here:
http://en.wikipedia.org/wiki/Receiver_Autonomous_Integrity_Monitoring

For the purposes of this discussion, the RAIM predictive function allows end users of the system to determine, well in advance, where areas of degraded signal coverage will occur, at what time, and the duration of the event as well. This would allow efficient planning of any shoot dependent on GPS signal coverage.

As far as the unit mounted on a flatbed and driven through a tunnel, or through the canyon that is downtown New York City, well, that’s another story. Clearly, the use of GPS is application dependent.

Another issue is Selective Availability. The satellite constellation is under the control of the US military. They have the ability to introduce ‘noise’ into the system, thereby degrading the accuracy of position in order to prevent ‘the other side’ from using the system to advantage. SA has been turned off for ten years, and as far as I know, Don Rumsfeld is busy writing his memoirs.:patriot:

So barring a major conflict involving the US and a credible outside threat, you can pretty much depend on cm-grade 3D accuracy of GPS.

As far as I can tell, GPS would be a valuable tool to provide gross alignment, the MEMS gyros, once given that information, refining adjustments from there. The more pertinent question though, already mentioned here, what degree of stabilization is required for, say, a 4K CMOS chip in motion, to acquire an image that is acceptable to the human eye?
 
Cliff Gill said:
maybe master thesis projects in producing a relatively low cost attitude and heading reference system?
Click to expand...

Finally found something. A master thesis from Luleå Sweden:

http://epubl.ltu.se/1653-0187/2007/086/LTU-PB-EX-07086-SE.pdf

Here's another one:

https://etd.sun.ac.za/jspui/bitstream/10019/380/1/BredenAFL.pdf

The references in these make it now bit easier. Have also found material published in the former Sovjet Union region. Perhaps a hint and explanation for the existence of this Ukrainian company selling gyrostabilized gimbals?
 
Some interesting information here. I found the video sensor input intriguing, as an additional aid to software generated stabilization.

Performance test results of a an intergrated GPS/MEMS inertial navigation package

http://www.navsys.com/Papers/0409001.pdf
 
Lauri and others found lots of great info! After going through some of the thesis it looks pretty challenging. I am exploring some existing gyro controllers at the moment and hope that they may work out with a custom gimbal.
 
Hi..
Interesting thread !!
I am not a RED user, but my neighbour has a RED ;)
Just joined so that I can make this post.

Perhaps I can offer something small in respect to the wishes expressed in the previous posts.

I have built up a small IMU device that can stabilize a single axis of a camera mount.
The device consists of a 3 axis MEMS accelerometer and 1 axis MEMS gyro
For 3 axis.. just use 3 units.

2 views of the device.

Picloc-TR_1.jpg
Picloc-TR_2.jpg


Its a modular unit, and for what it does, I think its very inexpensive.
The cost of it is around the same as that of a top R/C gyro.

If you'd like to see some videos of the unit in operation, and also some videos on how to set it up..
..search for the website address in the pictures above!
 
Last edited:
Hi..
The sampling frequency afforded by the microcontroller is about 350Hz.
Range of the Accelreometers is configurable from 1.5G Min to a Max of 6G.
However in real life, the G selection does not make a very big difference.

The drift from the gyros is not specified in the datasheet. However the noise is about 0.1º/s/root Hz.

In actual operation, with the unit mounted on the actual axis being actuated.... (ie.. the IMU is providing its own feedback for stabilizing(strapdown)).. there is no drift in Tilt and Roll applications, while the best drift in PanMode has been down to about 3 to 5º per minute.
 
Nice thread.

I had a dream quite a few years ago about a shot that was rather spectacular and very moving. I decided to go about creating that shot. I quickly realized nothing on the market could do it so i started designing a camera support. As one aspect i needed a modular camera and started researching cams and came across dvxuser the exact day Jim announced the Red One (everyone thought he was crazy). So i started designing the mechanic part of the head which is very near completion. (Red One was not suitable)

Has three axis all on the centre. Multi use (steadi cam head, crane, car, plane, hand held, under water etc etc..) Weight is minimal. Motors are minimal (exactly as you suggest Lauri) Very modular, No pullies, Gives tilt and roll to steadi cam. Can be packed up in a case measuring no more than 10" square x 7" wide. General cost will be around the $5000 mark. 50 degrees tilt in either direction and unlimited roll and pan.

i would really love to see this out there as soon as epic hits the market. I really believe it will help make some great shots that have yet been unattainable and its modularity makes it possible for some sweet additions to both motion control and one man ENG work, not to mention affordability and portability.

If a couple of guys are interested in collaborating, especially someone with programming/ motion control/ system feedback etc etc experience and someone with machining capabilities, please do let me know.
 
Count me in Jason. I have a guy doing machining work on another project, and will provide the aircraft and FAA paperwork for eventual field testing of the unit.
 
Jason Sinclair said:
If a couple of guys are interested in collaborating, especially someone with programming/ motion control/ system feedback etc etc experience and someone with machining capabilities, please do let me know.
Click to expand...

Jason, Thanks a lot for your post. Yes, I'm definitely interested in collaborating with you. The very idea of this thread was to create collaboration.
 
Sounds great guys. I will just pm Lauri some details as its his thread and there's no intention to hijack then hopefully we will get this show on the road.
 
Hello Jason. As far as the roll motion for use on a steadicam you might have to be carefull as it is the Allien, and A llien Revolution patent, if you plan to put this on the market. There are many electronics and sensor types on the market today most of them lack on some features to make this a ready to shoot system, even the Allien revolution has or had a lot of issues and only used roll cradle, it used to get confused when in big tilting angles and go crazy from some reports,and thats a very dangerous issue .
I myself have designed the Acrobat which works with a sensor and motor to correct horizon, still working on the sensor controller motor thing.
However the 5000 target seems to be quite a chalenge you will find that most suitable sensors with out electronics , programing, controllers can cost as much as 2000-3000 USD.I believe this project is something that will improve big time as it all depents in clever electronics and soon some company will come up with an of the box solution and in a very affordable price.
The mechanical part is easy.
Very interesting..
 
Maybe a stock package such as the Dynon SV-ADAHRS-200 could be utilized. From http://www.dynonavionics.com/docs/D180_Feature_ADAHRS.html :

"A set of three electronic MEMS-type accelerometers measure forces on all axes. Three electronic MEMS-type angular rate sensors (the modern electronic equivalent of "gyros") measure rotational rates on all axes. Three electronic magnetic sensors determine heading, and since Earth's magnetic field is also measured on all axes the heading is accurate with the aircraft in any orientation."

Dimensions are 4.71"W x 1.22"H x 2.61"D, weight is 8.2 oz.
List price is $1200.
 
Constantine Tyrintzis said:
, even the Allien revolution has or had a lot of issues and only used roll cradle, it used to get confused when in big tilting angles and go crazy from some reports,and thats a very dangerous issue .
Click to expand...

Constantine, Sorry to miss you by a few days in Greece last year. When you talk about problems with alien, can you elaborate? I can understand why titling the camera would totally stuff up that set up though.

Also do they have the patent on every conceivable roll mechanism on steadicam? How did they achieve that? That in my mind goes against all patent law but i'm willing to learn if i'm wrong. If they have achieved that, then well done to them because thats some serious patent application.
 
Jason Sinclair said:
Also do they have the patent on every conceivable roll mechanism on steadicam? How did they achieve that? That in my mind goes against all patent law but i'm willing to learn if i'm wrong. If they have achieved that, then well done to them because thats some serious patent application.
Click to expand...

what about if it's roll and tilt as well ;)

http://www.youtube.com/watch?v=u7aHsYHv2F8
 
You must log in or register to reply here.
Back
Top