PCMD estimations

I know for relative movements along NED we have moveBy, but this behaves clumpsy sometimes: It’s a long step, taking seconds, fails if fired too frequently, ends in hover, is hardly interruptible and so on.

I’m looking for a better solution for precise landing, Aruco backed up.

Thinking about PCMD, which works well in another use case, I was wondering, if there is kind of “rule of thumb” relation for a given tilt percentage value (for the four main directions in X and Y), which would allow to estimate, what distance in meter can be bridged under ideal conditions with one single shot :)?

I wanted to implement the same exact thing, precision landing on a tag. I ended up using PCMD too. I’m sorry I have not much added advice here, just “camaraderie” I guess.

The ANAFI Ai behaves correctly for moveBy according to the Olympe documentation. It smoothly moves from point to point without stopping. On all other ANAFI platforms I’ve noticed that moveBy hovers after completion and like you said, can be a little hard to control.

PCMD is working OK for me but it’s hard to control speed. The speed depends on the piloting settings, so you will want to adjust the settings and set them to a predictable value before you start joysticking the ANAFI around.

Thanks for confirmation. Yes, I’m also working with PCMD. I don’t find it hard to control speed, it is rather hard to control the result of an impulse - the final position. The main problem for me is that PCMD only accepts integer values as percentage of the max speed between -100 and 100. So the very interesting range between 0 and 1 cannot be controlled fine enough and needs an occasional variation of the max speeds in order to prevent overshoots/undercuts.

@ChristopherVo After a week of messing with PCMD I returned to moveBy and it works perfectly. Even at high winds.