20190807 Dev Meeting
General:
- Update from relevant previous calls
- Tagged: issues, PR.
-
JulianOes will discuss "MAVLink connection semantics" at next meeting. Expects to provide an overview to JamesP well before the meeting (next week).
-
JamesP to continue to push PeterB on reviewing the MAVLink bugs (Ollie) 339 and (Hamish) 338.
- UPDATED HamishW to address suggestions against his PR.
-
MERGED #1194 Opaque tunnel message
- Some concern that it might be abused, and questions about why no sequence number (now addressed by OlliW)
- Agreed to merge as WIP. HamishW to follow up.
-
#1165 Common Flight mode architecture next steps and docs
- JamesP - Flight modes messages/structure reflects PX4 internals, not necessarily best public interface for sharing and setting the information. JulianO agreed. James to comment the spec (Done)
- JamesP - would prefer to start with a small set of non-contentious flight modes. Commented here
- MERGED #1167 Add OpenDroneID messages - Would like to include in own dialect but mavgen support still pending. Approve merging in common.xml as WIP to allow further testing.
-
MERGED #1184 Reserving component IDs for private components - All agreed to merge. Action on Hamish
-
DONE #1179 Message SCALED_IMU - magnetic field - Agreed to use milligaus and add an extension for the battery ID. Action on Hamish
- Created MERGED #1199 SCALED_IMUn: x,y,zmag units from mT to mgauss and MERGED #1200 RAW_IMU: Add id extension to support multiple IMU to action this.
-
MERGED #1188 Clarify MAV_TYPE usage - Agreed makes sense. Merge. Action on Hamish
-
MERGED #1189 OPTICAL_FLOW.flow_comp_m_(x/y) units to m/s - James no familiarity with this message but says ArduPilot uses radians, incidating that they probably use the other variant of the message. JulianO says code doc for the component that uses it says velocity. Hamish says let's merge - clearly incorrect.
-
#1182 Clarify use of multiple batteries Agreed that SYS_STATUS/HIGH_LATENCY2 battery related fields should reflect single real battery, where battery is autopilot dependent, but expected likely to be settable using parameter. Action Hamish to update PR appropriately.
- Arrrgggg Further discussion/holes pointed out. Discussion continues.
See above!
HamishW, JamesP, JulianO