20230412 Dev Meeting

General:

• Update from relevant previous calls
• Tagged: issues, PR.

Notes

• Next week move meeting 1 hour later. AO hamish to get Ramon to move it.

• common.xml: remove MAV_CMD_ACK #1972

  • Agree merge. Isn't useful.

• Low Latency MAVLink Messages for VTOL

  • Telemetry data into audible signals over prioritized and low-latency voice channels in mobile networks.
  • Low latency, limited data rate.
  • Not clear what this offers over HIGH_LATENCY2 - what is specific choice being made here that needs this set. What about heartbeats?
  • Just a heads up today about the discussion.

• common - un-WIP gimbal messages #1973 - WIP on device

  • Mode message - what needs to happen?
  • Needs support for status connection info - see https://github.com/mavlink/mavlink/pull/1973#issuecomment-1502275438
  • Julian going to add a note that he is happy to review PR for a mode message if someone else creates it.

• Specification of board_version in AUTOPILOT_VERSION - #1968

  • AUTOPILOT_VERSION.board_version

  • What do last and first mean here:

    HW / board version (last 8 bits should be silicon ID, if any). The first 16 bits of this field specify https://github.com/PX4/PX4-Bootloader/blob/master/board_types.txt

  • PX4 seems to look like: 8: nothing (would be silicon ID) | 8: nothing | 16: board type Which would hint towards saying that the "first 16 bits" of the specs are the lower bits (LSB).

  • APM looks like MSB for board type. Nothing else seems to be used.

  • PX4 layout is (32) 8: nothing (would be silicon ID) | 8: nothing | 16: board type (0), so LSB. ArduPilot appears to be MSB.

    • ThomasB suggests must make whole field flight opaque data.
    • Maybe. Need to see what QGC says.

Not discussed:

• component metadata: add translation docs #498

  • FYI working implementation. Docs in final stages.
  • Request review by James last time - plan to merge.

• mavlink_msg_xxx_decode by mavgen_c does not handle extensions of message properly #684 - pymavlink bug fix

  • AO JamesP to see if we can get review from PeterB. HamishW to ask too

• message_definitions: add all_with_development.xml #1924

  • AO: JamesP to talk to PeterB to get movement on https://github.com/ArduPilot/pymavlink/pull/240
  • HamishW to ask too

• Added "shortest way" as an option for yaw direction in MAV_CONDITION_YAW #1962

  • We can merge this if there is a prototype that makes ArduPilot compliant.
  • AO HamishW to update.

Attendees: Hamish, ThomasD, JulianO