4
$\begingroup$

I'm designing an Unmanned Aerial Vehicle(UAV) similar in size to the RQ 11B Raven.

enter image description here

This UAV can carry color and infrared cameras to capture outside scenes in real-time. The quality of imagery that these UAVs relay to the operators is directly related to how much of information they can extract from the frames being captured in real-time. So far the captured video looks jittery and not appealing due to vibrations of the platform. I am encountering two issues here

  • issue of oscillations ; and

  • low frequency vibrations

Can you guys please suggest how UAV designers have managed to address these 2 issues to improve visual quality without compromising computational efficiency?

$\endgroup$
  • $\begingroup$ Mounting the cameras on shock-absorbing mounts (rubber bushings). Make sure the prop/motor is balanced. $\endgroup$ – Steve Dec 22 '16 at 14:11
  • $\begingroup$ The problem is commonly referred to as "jello". Here's an article about it: dronetechreport.com/… $\endgroup$ – Steve Dec 22 '16 at 14:12
  • $\begingroup$ @Steve: Thank You. I know the effect is called Jello in some circles however none of the materials I have come across explains how UAV designers do not compromise on computational efficiency while addressing the issues of vibration & oscillation. Large UAVs like Predator have a lot of computational power so these issues are not major issues, however for mini/micro UAVs striking a balance is extremely difficult. $\endgroup$ – CSinha Dec 22 '16 at 14:19
  • 3
    $\begingroup$ I'm not sure why computational efficiency is a problem, as this can be solved by mechanical means. $\endgroup$ – Steve Dec 22 '16 at 14:39
  • $\begingroup$ Yes, I realise it is possible. That's my question - the how part. $\endgroup$ – CSinha Dec 22 '16 at 15:14
7
$\begingroup$

First look at the frequencies involved. If your camera frame shakes so badly that the picture quality suffers visibly, I would suspect that your camera mount has an eigenfrequency close to whatever causes the vibrations (the engine, most likely). Maybe it is enough to shift the eigenfrequency of the camera mount away, either by changing the mass of the camera or by changing the stiffness of the mount.

It could also be that some structure other than the camera mount is in frequency neighbourhood with the engine and the camera is mounted near an antinode - maybe it is enough to shift the camera location a little.

Vibration problems are hard to diagnose from the little information you give, so all I can do is to guess the most likely cause of your troubles.

$\endgroup$
2
$\begingroup$

First, The low frequency dynamics that you mention probably comes from the movement of the aircraft, due to wind, gust, etc. one idea is to have a neutrally stable aircraft in pitching motion due to airspeed. This is not always possible but may help a bit.

Second, the high frequency distortion (shake) most probably is poor damping / installation.

Thirdly, to get crisp images cameras are built into gimbals, that actively suppress unwanted motions.

Lastly, postprocess video enhancement is possible, but for your application, probably this is not desirable.

And of course some cameras have video stabilization capability. That should solve half the problem.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.