Camera tilt problem

Discussion in 'Radian' started by Jim Swanson, Sep 22, 2012.

  1. Jim Swanson

    Jim Swanson Member

    Joined:
    Jul 13, 2012
    Messages:
    97
    Likes Received:
    6
    The camera tilts too far when flying and sees the red boom in the shot.
    I've tried adjusting the travel throws for the tilt servo on the transmitter but it has no effect.
    I have the tilt servo on the Elevator stick.

    Transmitter is a Spectrum JR 9303 2.4.

    Any ideas?

    Jim
     
  2. Tabb Firchau

    Tabb Firchau Administrator
    Staff Member

    Joined:
    Jun 28, 2012
    Messages:
    911
    Likes Received:
    472
    Hi Jim,

    When the Radian is stabilizing and there are no external pots there is no way for the system to know when it would see a boom. Your camera man will need to learn to compensate and look down a bit when accelerating forward in the beginning of a shot.

    The travel throws will just alter the rate of your tilt up or down.

    Best,

    Tabb
     
  3. Nick Kolias

    Nick Kolias Moderator
    Staff Member

    Joined:
    Jul 4, 2012
    Messages:
    140
    Likes Received:
    24
    Hey Jim, also keep in mind that boom position is constantly changing relative to the lens. So if you did limit the tilt so that a boom could never be in the shot it would needlessly restrict your frame for all the other times you could utilize more tilt up. So best to retain that range of motion and coordinate those moves with your camera operator.

    nick
     
  4. Jim Swanson

    Jim Swanson Member

    Joined:
    Jul 13, 2012
    Messages:
    97
    Likes Received:
    6
    Thanks for getting back to me quickly Tabb.
    So what does Tabb do when he flies solo with the Radians? How do you avoid shooting the boom?

    Seems like "something" told the tilt servo how far to go in each direction? Can this be changed in the program?

    This is from the manual. I'm not sure if this is relevant.
    ANGLE LIMITS:
    The stabilization system has programmed software limits of -110° to 10° on the tilt axis and -30° to 30° on the roll axis. Keep in mind these angles are relative to the ground reference, not the aircraft.


    I feel like we need to control the angle of the tilt some how or risk ruining a shot.

    Best,

    Jim
     
  5. Jim Swanson

    Jim Swanson Member

    Joined:
    Jul 13, 2012
    Messages:
    97
    Likes Received:
    6
    Hey Nick,
    I fly without a camera operator part of the time. On my 2-axis Cinestar I never tilt high enough to see my boom. This could be avoided in the MK system. I'm probably misunderstanding what you're saying, but I never want to see the boom in a shot and would be willing to accept less range of motion.

    Best,

    Jim
     
  6. Nick Kolias

    Nick Kolias Moderator
    Staff Member

    Joined:
    Jul 4, 2012
    Messages:
    140
    Likes Received:
    24
    Okay, understood. With the current firmware I don't believe the tilt angle limits can be altered in the setup software. Maybe this is something John can add to a future release. Let's see if he chimes in here.

    nick
     
  7. Jim Swanson

    Jim Swanson Member

    Joined:
    Jul 13, 2012
    Messages:
    97
    Likes Received:
    6
    John...what up?o_O
    Are you going to let Holger out-do you?
    Please show some American Pride?:D

    jim

    ps. the Radians work perfectly otherwise.
     
  8. Tabb Firchau

    Tabb Firchau Administrator
    Staff Member

    Joined:
    Jun 28, 2012
    Messages:
    911
    Likes Received:
    472
    Hi Guys, the Radian has no reference to the airframe so the "angle limits" you are referencing are with respect to the earth. As it stands right now there is no way for the Radian to know what the airframe is doing.

    At the extreme example imagine that you picked your multi rotor up off the ground and rotated it forwards 90 degrees(so the booms were vertical). The radian would keep the camera level even as it was looking through the booms.

    Hope this helps with understanding how the system works.

    Tabb
     
  9. Howard Dapp

    Howard Dapp Active Member

    Joined:
    Aug 18, 2012
    Messages:
    977
    Likes Received:
    111
    In simple terms (in my usual non technical explanation) MK is able to read potentiometer position feedback both from the TX and gimbal mounted pots therefore able to set limits on both. Radian does not rely on gimbal mounted potentiometer feedback therefore has no way of knowing tilt angles...without pots the system will not be able to set compensation tilt limits like MK. I do believe you are able to set tilt limit via stick input because Radian senses pot feedback from the TX.

    You'll eventually adjust by changing the way you fly...instead of quickly going into forward flight with an aggressive angle, you'll have to gradually build up speed to get there. I think the task of avoiding booms will probably fall more on the pilot than the operator in most cases. But then there are cases like moderate headwind that will call for a stronger tilt angle and obviously booms will drift into shots. The operator will have to be conscious and alert. I've had operators that were so focused on tracking a moving object that they totally missed the boom creeping in and out of the shot at the very very top of the monitor.

    Or you can lower the gimbal by an inch to gain a little more FOV..just some thoughts.

    Then again I may be totally wrong!:)
     
  10. Jim Swanson

    Jim Swanson Member

    Joined:
    Jul 13, 2012
    Messages:
    97
    Likes Received:
    6
    Thanks Tabb and Howard. I've pointed the camera down 15 degrees. That helps but I've lost my nice straight shot at the horizon. I dunno. Maybe if I had the added functionality of a 3-axis gimbal (I'm only 2-axis), the occasional boom shot might be less troublesome and could be worked around. Seems like this system puts too much stress on the pilot and operator to keep the boom out of the shot.
    With only 8 minutes of flight time, the last thing I'm want to hear is "boom in the shot, start over". Radian might not be for me.

    Again I love the rest of it.

    jim
     
  11. Howard Dapp

    Howard Dapp Active Member

    Joined:
    Aug 18, 2012
    Messages:
    977
    Likes Received:
    111
    It's the little things that some may think isn't so big but for others those little things are huge. I do agree that it would be nice to have a programmable limit for tilt compensation angle. There is absolutely no way a pilot and operator will be successful 100% of the time at avoiding booms in a shot BUT it is possible to be 100% successful if there was an adjustment to limit tilt compensation angle.

    It'll be most appreciated on those 9+ hour shoot days where you're doing multiple runs just to capture a single shot.. your whole body is tired and you're mentally exhausted. The last thing you want is to do a take over because of a boom and you definitely don't want to resort to cropping the footage.
     
    Flexacopter likes this.
  12. Dave Halton

    Dave Halton Member

    Joined:
    Sep 14, 2012
    Messages:
    112
    Likes Received:
    2
    So I guess there is no max roll limit on the roll axis either. I was trying to figure out how to stop my AV130 overshooting and stripping the gears or burning the servo out in the event of a roll over on landing or god forbid a crash. I would have thought that the Radion could be told what the angle limit should be before it stops driving the servos. Surely it knows when its level at calibration. I aint a software engineer but If DJI and MK and others can include it why can't the Radion software?
     
  13. Tabb Firchau

    Tabb Firchau Administrator
    Staff Member

    Joined:
    Jun 28, 2012
    Messages:
    911
    Likes Received:
    472
    Hi Guys, I agree that it would be nice to have an adjustable limit....the point I am trying to make is that it is impossible to do so right now without there being a separate Radian mounted to the frame of the aircraft. The MK board was able to do this as it used the external pots to define an upper tilt limit, but the Radian does not use the pots.

    I hope this makes sense. We are not trying to limit functionality here.

    Tabb
     
  14. Tabb Firchau

    Tabb Firchau Administrator
    Staff Member

    Joined:
    Jun 28, 2012
    Messages:
    911
    Likes Received:
    472
    Hi Dave,

    There is a max limit on the roll axis but it is with respect to the earth.....not the airframe of the craft. See my above post.

    The DJI and MK are able to do this as they use the external pots. Stabilization systems using External pots will never be able to perform to the level of the Radian due to issues with the potentiometers.

    Tabb
     
  15. Dave Halton

    Dave Halton Member

    Joined:
    Sep 14, 2012
    Messages:
    112
    Likes Received:
    2
    Aye Up Tabb.. OK I am dumb but surely a individual Radion knows what is level at calibration. So why cant then a limit be set to 30 degrees either side of level?

    If it cant it can't but can not quite get my head around why not.

    Dave
     
  16. Tabb Firchau

    Tabb Firchau Administrator
    Staff Member

    Joined:
    Jun 28, 2012
    Messages:
    911
    Likes Received:
    472
    Hi Dave, the Radian certainly can tell when IT is 30 degrees each side of level, what it cannot tell is what the airframe above it is doing.

    For instance imagine picking up your multirotor and rotating it 90 degrees, the Radian would still be level but the aircraft would be 90 degrees from level.

    Does that make sense? The key point is that the Radian only knows it's angle, not the airframe's angle.

    Tabb
     
  17. Dave Halton

    Dave Halton Member

    Joined:
    Sep 14, 2012
    Messages:
    112
    Likes Received:
    2
    Ill sleep on it ;)
     
  18. Howard Dapp

    Howard Dapp Active Member

    Joined:
    Aug 18, 2012
    Messages:
    977
    Likes Received:
    111
    Tabb, your creation is no doubt great and we all love and respect you for it:)

    Just thinking out loud here, would it be possible for the pan Radian to also act as a "frame" mount unit? It would obviously stay mounted on the gimbal as it is now and not the actual frame, but since the pan Radian is decoupled from the gimbal's tilt and roll axis it could act as the position sensing feedback unit since it tilts and roll with the frame? It would be possible for it to give feedback to the roll and tilt Radians on the pan axis/frame tilt & roll angle. This actually sounds possible..
     
  19. Dave Halton

    Dave Halton Member

    Joined:
    Sep 14, 2012
    Messages:
    112
    Likes Received:
    2
    Doh! just got it! o_O
     
  20. Tabb Firchau

    Tabb Firchau Administrator
    Staff Member

    Joined:
    Jun 28, 2012
    Messages:
    911
    Likes Received:
    472
    Howard, just to be clear the Software side of the Radian is 100% John's baby. I was just lucky enough to get to use it first ;) I have been begging John for years to develop something along the lines of the Radian as I was previously using his CARVEC system in a similar role and it worked excellent.

    Good idea on using the Pan Radian....the other issue for me is if that even if you impose a 'limit' on the tilt the shot is ruined either way. You either hit an artificial limit and the shot looks bad, or see a boom and the shot looks bad....it is really a tough problem with multi rotors. I think 2 person teams that work together frequently will learn to work around it without problem. The only time Hugh and I get a boom in shot these days is when I do an agressive blast off or some other stupid maneuver. Like you said though I have learned to ease in and out of forward moves to keep the boom out of his way.

    Another technique we use is to use an extreme burst of power at the beginning of a shot (before the shot will start) to get the multi rotor up to speed and then you can ease off the tilt angle and avoid the boom.

    Tabb
     

Share This Page