20181212 Dev Meeting

Agenda

General:

• Update from previous calls
  • AI @ALL - review RFC in #861 "Local and body frames" for next meeting.
• Tagged: issues, PR. A few choice items:
• Pymavlink#237: message CRC checksum is wrong when a mavlink_message_t is converted to a sending buffer - quick update.

HamishW

• We discussed extending STATUSTEXT. Can't find any issue or solution agreed. What are thoughts on this?

CHECK Status

• @JamesP and @MichaelS (or Lorenz/Thomas) to engage with Local and body frames and add comments. Discuss next week.
• @WickedShell File format discussion - Attempt to re-engage in next two weeks.
• @JulianOes to review and respond to message CRC checksum is wrong when a mavlink_message_t is converted to a sending buffer

Notes

Call 1

• @HamishW: Next meeting deferred. We will meet again Jan 9, 6pm AEST.
• @JamesP:
  • Pymavlink has had some work this week in mode handling and multi-vehicle link support. There is still some work to do in pushing through behaviour to DroneKit and testing.
  • Pymavlink#237: message CRC checksum is wrong when a mavlink_message_t is converted to a sending buffer - still no update on what needs to happen to progress. Will chase.
• HamishW: Local and body frames
  • Needs key players to discuss. Should also keep Rover in mind.
  • AI on HW to try get this acted on
• HamishW STATUSTEXT discussed previously but no action
  • JamesP - what is driver for this? Why is it being abused?
  • HamishW - we're not sure it is being abused or how to stop that. We are sure that 50 chars is too little.
  • Proposed solution is to add extension field that is additional information. GCS should construct total status from both messages, if present. This is more hassle for users but probably best compromise.
  • AI HamishW to propose message extension as PR. Invite PeterBarker to contribute

Tagged issues/PRs:

• common: Add a relative jump command #1033
  • We're happy with the name and proposed implementation
  • Discussion of what to do in case where mission has multiple tag of same type. Ideal is disallow, but easier said than done. Decision was to recommend jumping to first discovered in greater numeric mission order. Also recommend GCS disallow. during mission definition.
• Clarify Global Frames #1012
  • HamishW - Propose merge as is editorial not content change.
  • JamesP: Are points are relative to "Home" or EKF origin.
  • HamishW - Does not matter because a)Not every system uses EKF, b)This PR is not aiming to redefine message, just make easier to use. c) If people want to redefine the message, that is a more substantial PR.
  • HamishW - Plan to merge tomorrow if no comments.
• RC channels RSSI definition does not match prior art #1027
  • RSSI field is byte (max 255 decimal). On many systems defined as 0-255 full range. In spec defined as 0-100 as %.
  • Generally % easier for GCS but reduces resolution that might be useful for antenna tracker.
  • Given PX4 is outlier and PX4 implementation appears to be outlier, proposal is to revert spec and PX4 to same as other systems. AI on HamishW to follow up.
• MAVLink battery information #988
  • Discuss whether we can have shared power/battery message. Not clear, but looks like we should have different message for power rails/power monitoring.
  • Agreed that we need to meet battery->autopilot->GCS scenarios, and also bridge SMBus and UAVCAN smart battery information.
  • Action to find out what UAVCAN bridging message might need.
• Proper deprecation of DO_DIGICAM_CONFIGURE and DO_DIGICAM_CONTROL #967
  • Already agreed in dev meeting, but stalled (by spurious question from Hamish about tooling)
  • re-engage with Amlicar to get this in within week. AI on HamishW
• Add rover message to common.xml #1032
  • looks like progressing well, but might need some pushing along.
• add CELLULAR_STATUS message to report cellular network status #932
  • This looks relevant. No other messages cover anything but latency.
  • Not much expertise on this in group. AI on Hamish to see if we can get some thinking on this from Auterion.

AI:

• JamesP - Pymavlink#237: message CRC checksum is wrong when a mavlink_message_t is converted to a sending buffer - find out ETA/help needed to progress
• HamishW (Done (merged)) - merge Clarify Global Frames #1012 if no further comments.
• HamishW (Started) - RC channels RSSI definition does not match prior art #1027 - attempt revert spec to previous version. Will require PX4 and QGC changes.
• HamishW (Doing) - MAVLink battery information #988 - action to engage Pavel for UAVCAN info and generally seek consensus with community (including JeffW)
• DONE by Amlicar : HamishW - get - Proper deprecation of DO_DIGICAM_CONFIGURE and DO_DIGICAM_CONTROL #967 - merged.
• HamishW (Doing) - Add rover message to common.xml #1032 push towards conclusion.
• HamishW (Started) - add CELLULAR_STATUS message to report cellular network status #932 - Kick it off again and seek advice from Auterion.
• HamishW (Doing) - common: Add a relative jump command #1033 - add notes from meeting to issue. Attempt to get general agreement on this in coming week.
• HamishW (Doing - Allow bigger STATUSTEXT messages #1038) - STATUSTEXT propose message extension as PR.

Attendees:

HamishW, JamesP, JulianO, MichaelDB, ted!