I have dual Here 4 units but havent tried this yet. Is anyone successfully using this feature? Any issues that you can share on why I might not want to use it?
This post is interesting but makes me nervous that it may not be ready for primetime:
Also, the GPS/Compass landing page mentioned “a version of firmware for the CubePilot HERE 4 DroneCAN RTK GPS is available here that includes evolving ArduPilot improvements.”
Should I be using this firmware on my Here 4 units rather than the firmware that it came with/updates through mission planner?
Im currently running the latest version mission planner will install when I select update firmware, which is 1.14.D42D3330:
I’m not aware of the latest developments on the subject, but I know you should be very cautious updating Here4 firmware due to complications with this config in the past.
The rule is : Always update the ap_periph firmware on the unit to the latest, since bugs are fixed. BUT newer update the U-blox firmware on a here4 since their fw is a custom one which includes moving yaw (which is not a standard feature of the ublox chip in the unit).
@Eosbandi but what’s the right way to update that firmware? Up until today I would go to the dronecan tab in mission planner, select each Here4, and then choose the update button. It applied the latest 1.14 firmware
You should update it from mission planner, they are maintaining a separate branch, based on the ap_periph… the latest version is 1.14.4 ( [d42d333]) and MP downloads it from their servers.
As for the moving yaw setup, is anyone using it and have pros/cons as to if it’s reliable and better than a compass? If I lose a GPS will the system revert back to using 1 GPS for position and then the onboard compass?
Thank you all. I set it up and it seems to be working so far, but I have only done some light hover testing in my backyard. It was super easy to get running, just follow the docs. It does take a few minutes for my hexacopter to get a position and arm though now. I suspect it’s needing to get a more accurate GPS position than it would normally need so that the moving baseline works?
Yes, it may take a little longer for the fix to improve enough for the yaw solution to be valid. A few minutes seems longer than “normal,” but that’s very subjective (and subject to the environment as well).
Be sure that you’ve measured the distance between the antennas as accurately as possible and applied the parameters accordingly. It doesn’t have to be perfect, but the farther from reality it is, potentially the longer you wait (depending on the direction and magnitude of the observation error).
