ArduPlane with auto ship-landing — a LUA script continuously updates HOME from a moving beacon (over a separate MAVLink link forwarded through MP).
Observed behavior
When SYSID_THISMAV on the connected vehicle is changed during a session, the home marker on the FlightData map freezes at its previous position and no longer follows further home updates. - The vehicle’s actual HOME continues to update correctly (verified via RTL). - HOME_POSITION packets reach MP and update its internal state. - Only the on-screen home marker on FlightData is stuck.
Workaround
Performing “Read WP” once in the FlightPlanner tab after the sysid change restores normal marker updates.
var wps = MainV2.comPort.MAV.wps.Values.ToList();
if (wps.Count >= 1)
{
var homeplla = new PointLatLngAlt(MainV2.comPort.MAV.cs.HomeLocation.Lat,
MainV2.comPort.MAV.cs.HomeLocation.Lng,
MainV2.comPort.MAV.cs.HomeLocation.Alt / CurrentState.multiplieralt, "H");
// ... home marker is created inside this block via WPOverlay ...
}
in fact, not every operation uses auto mode (missions); some operators may fly primarily in GUIDED mode, especially for maritime operations.
Also, SYSID_THISMAV may be changed before flight as well, and the issue still occurs in that case. The important point is that the behavior is reproducible whenever the system ID changes for any reason during the same MP session.
There are scenarios where users intentionally change system IDs because some backend/web platforms distinguish vehicles by MAVLink system ID, and that is also our case (although we are gradually moving toward hardware-based identifiers).
We have also encountered real-world cases where inexperienced operators confused SYSID_THISMAV and SYSID_MYGCS and configured them to the same value. As a result, the vehicle could no longer properly receive GCS heartbeats, which triggered a failsafe condition. However, because the operator misunderstood the warning and continued flying anyway, the situation effectively became one where the failsafe mechanism was being ignored. Consequently, when the actual communication link was later lost, the aircraft did not trigger RTL as expected.
For that reason, we do not consider SYSID-related issues to be merely theoretical edge cases.