Cannot configure Moving Baseline (CubeOrange + 2x Here3+)

I currently have a VTOL aircraft (ArduPlane V4.4) with compass interference problems. The aircraft has a CubeOrange and two Here3+. I have thought about using the Moving Baseline as a positioning method. Following the tutorials has not been effective (thanks @Yuri_Rage for the great contribution to the community). I was hoping that someone who has had a similar problem has come up with the solution. The list of parameters I have configured to try to get the MB up and running is as follows (the rest of the parameters are configured by default or with a configuration that previously worked, obviously without MB mode):

  • GPS_AUTO_SWITCH = 0
  • GPS_DRV_OPTION = 8
  • GPS_AUTOCONFIG = 0
  • GPS_BLEND_MASK = 7
  • GPS_MIN_DGPS = 50
  • GPS_MB1_TYPE = 1
  • GPS_MB1_OFS_X = 0.5
  • GPS_MB1_OFS_Y = 0
  • GPS_MB1_OFS_Z = -0.05
  • AHRS_EKF_TYPE = 3
  • EK2_ENABLED = 0
  • EK3_ENABLE = 1

Both Here3+ are connected via CAN on different Cube can ports. I don’t know if the implemented hardware is able to achieve what I am looking for, so I don’t rule out that this is the problem.
From total ignorance I think it is possible that I have not configured the RTK correctly or not even configured it at all. The Here3+ manual indicates the steps to follow for one ground and one on-board station but I have not found anything about configuring for two on-board GPS. I would appreciate if someone could check the parameters and detect the error because I am out of ideas. I have never seen anything like “float rtk” or “rtk float fixed” on the screen, only “3D fix” on both GPS with the non-MB configuration. Currently, with the described parameters I have 3D fix on both GPS but GPS yaw is still 0.

If I have forgotten to add any information please let me know and I will update the post.

Other doubts that I have and I don’t know if they interfere in any way in the configuration:

  1. When setting the parameter GPS_AUTO_CONFIG=2 in the Cube, inside the GPS the GPS_TYPE changes to 17 and 18, but according to the source code (link, lines 125-126, 130-131) these parameters are not for DRONCAN (and the ones in my device are CAN) so I think the correct option is to overwrite those parameters with the values 22 and 23.

  2. Although, in principle, the MB configuration I think is correct I don’t know if there is any extra configuration to do to enable RTK injection or something like that.

  3. There are two parameters GPS_MB1_TYPE and GPS_MB2_TYPE, but I think that with only two GPS I should configure only GPS_MB1_TYPE (and then introduce the offsets between the antennas, for this I followed this guide from HolyBro link).

I don’t know if I’m completely wrong or I’ve configured it all wrong from the beginning or not even if the hardware I have is compatible with the functionality I’m looking for. With all the context maybe it’s easier if someone who has faced a similar situation can point me in the right direction or where to look further.

Follow these instructions:

https://ardupilot.org/copter/docs/common-gps-for-yaw.html#dual-dronecan-gps

Do not use that Holybro doc. It does not apply!

I will try it again (my first steps were with those instructions). Do you know if the here3+ have some kind of incompatibility with MB? I have read that some people say they are not suitable for this job, hence my doubts about the hardware I use. Ty for your input.

They aren’t my choice for the job, but they should work. You have multiple errors that can be solved by following the linked documentation carefully (many of your parameters do not match the docs, and your modules should be on the same CAN bus).

The auto configure routine works properly. Stop trying to override it.