Dist_bottom estimation diverges significantly from rangerfinders output

Hello, I’m back after causing more trouble!!

The range finders output and dist_bottom estimation differ significantly. I am using the Holybro H-Flow optical flow sensor as the rangefinder. On the ground, the rangefinder reads around 0.08 m and the dist_bottom records just above 0. I’ve attached the flight review and some plots to show what I mean. The flight review shows some more problems, but most of them come from issues with the rangefinder.

Here’s the catch: dist_bottom_sensor_bitfield is 1 which corresponds to rangefinder so supposedly the rangefinder is being used in the estimate for dist_bottom. cs_rng_terrain is 1 for the whole flight so it should be being fused for terrain estimates (which to my knowledge is what gets used for dist_bottom). The dist_bottom_valid flag is 1 for the majority of the recorded flight especially on the ground. dist_bottom_valid flag is 0 however when it starts to take off.

To me, this all points to that the dist_bottom should be equivalent to the range finders readings for the most part. I tested this in simulation and got the result I expected.

What would the reason be for dist_bottom and the range finder not being even close to eachother?

Flight Review: https://review.px4.io/plot_app?log=11f10363-3c39-475c-9c38-4c3068dc0ac3

Plot:

Your rangefinder got stuck reporting 3.3m

The terrain estimate (dist_bottm) is a function of more than just the rangefinder. The flow velocity itself can be used to estimate terrain, since it can be correlated with body velocity which allows the calculation for distance to ground.

Sorry for the late reply.

I found that my rangefinders output was hovering just barely below the height threshold parameter for fusing it into dist_bottom estimates. I put a bit more slack into the parameter and the dist_bottom estimates seems fine now!

This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.