Video Stabilized Gimbal Project

Hi all,

I need to do a Video Stabilized Gimbal. I think it’s possible with Arduino.

I need to control two DC motors (pan and tilt) with PWM input. I will need optical encoders for the position. And also an IMU unit for the stabilization. It’s possible?

Can anyone guide me? This is all new for me. Your help will be very appreciated. Thanks.

If this is all very new to you then I think this is a far too ambitious project.

What you are trying to build is a closed loop servo system on two axis. This is the sort of project you would give to a final year degree student in computing or engineering. Defiantly not one for a beginner.

I can pay to someone to do it for me. Anyone have capacity to do it?

I'm Aeronautic Engineer. I study 5 years in University. With the right guidelines, I can do it. I just need some help.

Hi again, I have now the Arduino and I'm starting this project. I understand now how to read the PWM input from the RC receiver. It's a beginning.

Now I'm trying to understand the IMU iworking and Kalman Filter teory.

I need also to choose an IMU unit, the dc motors and the encoders.

For the encoder I'm thinking about this absolute magnetic encoder:

http://www.rotaryencoders.com/products/encoders/absolute/rotary/kit/mae3

Have output in PWM or analog. What do you think? Good choose?

Any hint about good dc motors? They need to be small and light. The camera it's not heavy. About 400grams.

The weight of the camera is not important, the expected inertia is. You can move a heavy camera slowly or a light camera quickly and they'll have the same inertia to overcome. What type of motion are you trying to counteract? Hand tremors? Road vibration? Slow variations in angle? The higher the frequency and magnitude of the motion you are counteracting, the harder it will be. A 400 gram camera is not light at all if you're trying to move the entire thing for stabilizing a lot of vibration. The motors will probably need to be about the same size as the camera.

It's for a UAV camera gimbal. Unmaned Aerial Vehicle.

Any help please?

I'm thinking use this Sparkfun IMU 5 Degrees of Freedom:

http://www.sparkfun.com/commerce/product_info.php?products_id=741

5 DOF? Doesn't the platform yaw?

Only Pan and Tilt.

Pan and tilt...or yaw and pitch?

For gimbals we use to talk about pan and tilt.

I'm thinking here about this. Please, give your opinion.

If we have an IMU in the video gimbal, we will need also the encoders?

If this project is for an unmanned aerial vehicle, why on earth would you want to build a lot of hardware into the plane? All that extra weight.. Not good! =P

I assume the plane transmits video to a computer on the ground, right? Or is your plane so large it carries it's own recording equipment? If it transmits images to the ground, I'd just as soon go for a software solution. Video-smoothing software on the receiving computer could do the job of all that heavy, expensice equipment just as good, if not better.

Good luck with your project!

In ground I will use in addition a video stabilizer hardware.

You would need a feed (low bandwidth) to see what you are aiming your camera at, but it surely would only be reference video, not archival quality. If you are shooting HD this becomes even less of an option because of the increase in bandwidth. I thik Tapeless capture (P2, acdhd, mpeg2) would be the way to go, then ingest the video upon the return of your aircraft.

But back to the stabilization, there is actually a lot to consider here.

  1. Are you just wanting to stabilize the motion, or will you also steering the pan/tilt mechanism?
  2. Is this for a heli or plane?
  3. Think about adding ability for roll too.
  4. You need to look into RC hobby areas for gyro stabilization. Many heli mounts (life sized) use x,y,and z gyros. There are gyro systems for RC devices too. But you also then need to think about if they will be rate locked or heading locked + How will you compensate for the difference of planned change in motion (pilot input) verses unplanned (wind, thermals, birds, small artillery).

If you want to do all this, as previously mentioned, it is going to add a lot of weight. How big is your aircraft?
What size is your camera.
Is the aircraft gas or electric? I ask this because now you are lugging a bunch of weight, so your flight time just dropped significantly. If you are electric, extra batteries add extra weight too. All of it is doable, but the cost will also start to add up real quick too. how high are you wanting this to go? if you go uber high, then you will need more glass, which also is more weight.

1. Are you just wanting to stabilize the motion, or will you also steering the pan/tilt mechanism? Also steering the pan and tilt.

2. Is this for a heli or plane? For a aircraft.

  1. Think about adding ability for roll too. No.

4. You need to look into RC hobby areas for gyro stabilization. Many heli mounts (life sized) use x,y,and z gyros. There are gyro systems for RC devices too. But you also then need to think about if they will be rate locked or heading locked + How will you compensate for the difference of planned change in motion (pilot input) verses unplanned (wind, thermals, birds, small artillery). I will use 3 axis accelerometers and 2 axcis gyroscopes.

The aircraft will be gas. The camera is a Sony block camera.

I don't know what will be better for this. If DC motors or Stepper motors. Any ideia?