EN CN
  • Hey everyone,

    I am looking into building a 3 axis camera gimbal. This would be for my 4th year project in university. I am gathering as much information as I can before I get started. I am in third year now (half way through).

    I am confused as to where the Inertial Measurement Unit (IMU) goes on the gimbal. I thought it might go where the yaw axis is, since that is where most 3 axis gimbals are held from and that is where all of the movement happens. However I have seen in many videos that people have them on the underside of where the actual camera is mounted. I don't understand why this is, isn't that where the movement is supposed counteracted, and by that I mean, isnt the point of a gimbal is to minimize camera shake and movement?

    Any input is greatly appreciated.

    TL;DR: Where does the Inertial Measurement Unit go on a gimbal, and why?
  • IMU measures angular velocity and linear acceleration.

    By default when stabilizing the camera does not move (angular), the goal is to stabilize the angular movements.

    So on this regard it does not make difference where around the camera the IMU is placed. If one part of the camera rotates, so does the other and same with acceleration.

    But there is some non ideal elements on this. as it is your project, you need to do the thinking ;)
  • Thank you for the input!
  • Benkovy, the ideal placement for the IMU is at the intersection of all the axes on the same piece that the camera is attached to. I have found that the most critical axis to try to align with is the YAW. That is, I try to position the IMU directly in line with the YAW axis, underneath the camera tray. Does that make sense?

    I disagree with Garug about IMU placement not making a difference. As Garug has pointed out, the IMU measures angular velocity via gyroscope, and linear acceleration via accelerometer. So, while the gyroscope measurements will be the same regardless of where you place the IMU, the accelerometer will experience more linear acceleration the further away it is from the center of that axis.

    So I repeat, the ideal placement for the IMU is at the intersection of all the axes, but in reality that is where the camera mounts. So you have to "optimize" for a single axis, and that is why I try to place the IMU in line with the YAW axis, because that is usually the axis with the most mass attached to it.
  • "...Garug about IMU placement not making a difference. " not so, as I mentioned "But there is some non ideal elements on this. as it is your project, you need to do the thinking ;)"