June 03, 2020
- Status update by component/project
- Roadmap, and Release discussion
- Community Q&A
- In-Depth discussions
Meeting ID: 946 175 205
Join using your mobile/desktop
Call in using your phone, find your local number: https://zoom.us/u/aetSYKbiMF
OS / NuttX
Issue fixed in NuttX upstream related to DMA, it was the root cause for the durandal not working. There’s chatter about a new version 2 months out from NuttX master.
FMU v6x bring up changes are work in progress and should get merged soon. Probably next week.
New drivers for BMI88 BMI55:
The new drivers should move into the manufacturer folder and at some point when they’re well tested we can remove the old ones.
Issue fixed with the yaw estimator when the unobservable gyro bias gets too high on the ground which results in spinning heading and no convergence:
Solution for the reported altitude problems with 1.11
Proposal to remove an unexpected feature. Follow up needs to be to remove maximum altitude from the land detector message.
New releases with bugfixes/small improvements:
Python reference docs: http://mavsdk-python-docs.s3-website.eu-central-1.amazonaws.com/
GitHub pages migration of all the docs. MAVLink DNS works fine now. QGC still has a HTTPS certificate issue @LorenzMeier is looking into it.
Question about styles between the docs. @LorenzMeier is asking if it makes sense to distinguish dev guide and user guide. There are multiple people in the call agreeing that it is quite confusing to have content distributed over both. We have to talk to @hamishwillee but the consensus is that merging the two guides could avoid a lot of confusion. It would probably be structured to start out with usage and configuration instructions and go more in depth after that is covered. E.g. high level first install QGC and fly and then install toolchain.
MAVROS / RTPS / ROS2
There seem to be issues with certain CI tests catkin, offboard attitude. @dagar proposes to comment not passing tests out and assigning issues to the right people to fix them since we cannot start dragging prs out because of the repeatedly failing tests. He’ll do that after the call.
MacOS toolchain was broken and now it’s fixed. Homebrew simulation package maintained by osrf was broken for some weeks and was fixed.
For UAVCAN v1 we get close to something mergeable that we can start using.
NXP successfully tested UAVCAN v1 implementation with their BMS and RDDrone S32k board.
There’s various reviews and tests going on for the message definitions. The idea is to have them in the dronecode organization and use them in PX4.
Summary blocking for v1 public beta:
- Define repository
- Define message namespace
- Pick message IDs
There’s a dedicate call on that every Tuesday. Join next week if you’re interested.
See Dronecode calendar: https://www.dronecode.org/calendar/
Roadmap, and Release discussion
Blocking altitude issues were solved, see Estimation. @dagar solved a lot of small blocking issues. The release candidate is around the corner ICM issue, ESC calibration are the only possible remaining blockers he can think of at the moment. We’ll check those and
brianphilips: Suggestion to simplify the adoption of MAVSDK by providing issue templates for feature request and bug reports like on PX4. This is a great idea and it’s highly appreciated if there was a pull request for templates.
Errata and Feedback
Let me know below if I failed to capture anything the right way, and if there are any updates, or you have feedback on the call format.