Radar behavior is too aggressive when you fly over trees
We have also tried the lidar and it has the same problem.
I had hoped that the radar would fix it because somewhere it says that it penetrates the vegetation, but not at all.
The issue is that you need some kind of filtering. At least be able to modify the speed of ascent and descent. But the parameters WPNAV_SPEED_UP WPNAV_SPEED_DN don’t work for this because it seems that when you use the radar you only use the height that comes from the radar.
I agree, this is the biggest problem with Arducopter‘s rangefinder terrain following. Some kind of algorithm that averages the measured height during a short distance would be nice.
Or, maybe it could be a approach, if there would be a posibility to adjust the climb/decent speed during terrain following (independend of WP_NAV_SPEED)
Thank you for the hint. I tried different values with this parameter in the past. But it was a litte bit mysterious to me. Are here any units, or any further explanations?
I’m using terrain following with rangefinders under very different circumstances for a while now. For example, we are flying above whine yards with terrain differences of more than 80m within a distance of 200m. I know the documentation quite well, I think. But I never find a real explanation of RNGFND_GAIN.
It is a fact that the rangefinder based terrain following ist quite aggressive. I’m not the only user who would like to have the possibility to “soften” the terrain jumps a litte bit. Maybe RNGFND_GAIN is the right hint, but I do not understand the exact function of this parameter and its values.
it is a nanoradar NRA24, it has a special firmware and emulates the “ulanding”. it seems to work great.
We have done some quick tests with RNGFND_GAIN and it seems to smooth the behavior. Our drone is big and it’s scarier than a small one when it sinks. I think RNGFND_GAIN will work if we become familiar with it and adjust it for each scenario.
Still testing!