I I wanted to prove to myself that I could program an Arduino Development board to function as a flight controller. Pretty much the same algorithm can be used to control a camera gimbal to keep an action cam steady on a wobbling quadcopter. But since I got the quadcopter to fly already, I did not feel the need to prove myself all over again.
HAKRC 3-axis Gimbal to put my cheap action cam upon. I have a more expensive Drift HD Ghost, but that one weighs about 200 grams, while my cheap no-name GoPro clone only weighs 56 grams... S the choice is simple. It will not render super quality images, with 1080p at a very high compression rate, but it will give a nice idea of the possibilities.
So I just bought myself a
HAKRC 3-axis Gimbal to put my cheap action cam upon. I have a more expensive Drift HD Ghost, but that one weighs about 200 grams, while my cheap no-name GoPro clone only weighs 56 grams... S the choice is simple. It will not render super quality images, with 1080p at a very high compression rate, but it will give a nice idea of the possibilities.
But, just simply spending money on a gimbal doesn't get you there. There's a whole brunch of parameters you can play with to get the response you want from each of the three axes. It is hugely important to make sure your action cam sits on the gimbal in perfect balance, so that the motors do not have to constantly correct the imbalance on top of all movement. I've got it to a point where my camera does what i want when i move the quadcopter by hand. But when I fly, the gimbal switches stabilization modes at some point during take-off. It switches from a so-called Hold-Hold-Pan mode to Pan-Pan-Pan mode. The words stand for the response in the three principle axes: tilt, roll, yaw. I want the camera to hold it's orientation in space in the tilt and roll directions, and yaw smoothly ("dampened") with a yaw movement of the quadcopter. The mode it switches to is as if it is stirdily mounted to the quadcopter frame.
At first I thought it might be because of a voltage drop when throtteling up. But after trying with a separate battery for the gimbal alone, the effect still was there. So it probably has to do with vibration causing an overload of corrections.
This is pretty much where I stand right now with this quadcopter. As you can see from the video, I haven't modified (or flown) it since last winter -over half a year ago.
Why? You'll find out in the next post ;)
Comments
Post a Comment