20220525 Dev Meeting

Agenda

General:

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

Agenda

• Clarify and cross reference home position everywhere - OK to merge and ignore whitespace

  • Hamish - whitespace is ugly but non breaking. Allows better docs rendering. Non breaking. Can be improved in generator.
  • JamesP merge - this has been tested.

• RFC 0018 - Improved Battery Status Reporting #19 + WIP smart battery messages #7 (dronecan)

  • This is basically aligned. A little churn, but probably close enough to prototype.
  • General feeling is that development.xml can merge and basic prototyping be done in the coming weeks.

• WIND_COV - https://github.com/mavlink/mavlink/pull/1845 revamp to WIND_INFO and fix up the field names?

  • AO Julian - Fix up the descriptions with mention that none of the values are covariances. Can then be merged.
  • AO HamishW to see if WIND could reasonably be copied across and extended as a less confusing replacement.
    • Yes, but probably not worth it. The main difference is that WIND has a single vector for direction, while WIND_COV breaks the speed up by direction. We could add to WIND but the fields would be extensions, so we'd loose the benefits of the type check for little benefit.

• SYS_STATUS battery - path for deprecation?

  • Should battery fields be deprecated.
  • James P. The fields work for the vast majority of cases (single battery vehicles), and forcing a change at this time does not add value.
  • AO HamishW - add some docs about the limits in the SYS_STATUS mesage. - DONE https://github.com/mavlink/mavlink/pull/1850

• HOME_POSITION - quaternion can have an invalid value (following issue in HOME_POSITION Q)

  • Not possible to provide "just yaw" and indicate other fields in valid.
  • Therefore should either provide all information or indicate all of it is invalid.
  • AO HamishW. Update docs to indicate that. - DONE.

• FYI/Heads up add joystick to metadata type #1848

  • Not clear what the intent is. Need more info to evaluate.

• Mavlink C header length - looks wrong - can we get some eyes?

  • Probably the headers aren't being used.
  • AO Hamish to update them and see if anything breaks :-)
  • Reviewed. Headers are fine. Misunderstanding

Attendees: Hamish, JulianO, JamesP,