June 17, 2020
Agenda
- Status update by component/project
- Roadmap, and Release discussion
- Community Q&A
- In-Depth discussions
Join Meeting
Meeting ID: 946 175 205
Join using your mobile/desktop
Call in using your phone, find your local number: https://zoom.us/u/aetSYKbiMF
Component update
System Architecture
@dagar lost sleep over wrong external mag rotations that lead to crashes before. He implemented an automatic detection for the external magnetometer based on the knowledge of the rotation of the internal magnetometer of supported flight control boards. It compares the mags during calibration and chooses the rotation for the external mag with significantly the least error. There are some safety guards to prevent setting the rotation based on disturbance while calibrating. Testing and high level review are welcome.
OS / NuttX
@david_s5 Socket CAN was merged in upstream NuttX. It would be worth to get up to date again. He’s starting to work on it. We need to take care that we don’t have any downstream changes that we didn’t contribute and would lose when updating. David is going through the commits.
@dagar Is planing to finish a hardware test rig with a lot of boards to make the update testing quicker.
Support for ARM GCC 9 is almost there and would also be useful for that. Let’s make a final push to get all board working since we already have testing for part of them.
Driver
@dagar did a lot of work on various sensors:
For magnetometer testing a fake mag driver
Commander
The power redundancy check fails when bench testing powering over USB. We should
Estimation
@dagar made updating the magnetic declination lookup table automated using a script.
@dagar has a pull request to move PX4 dependent things from ecl dirctly into the PX4 repository. There are no objections but we should check with @LorenzMeier why they were put there at the beginning.
VTOL / Fixed Wing
We should have a combined project board to track all the open pull requests and issues and bring them all the way through. @sfuhrer is on board with that idea.
Multicopter
@dagar 's idea is to only ever use the altitude above home close to home such that it has a chance to be correct.
@igalloway is happy with the ground slow down since he knows it’s a feature. Documentation and notification to the UI would help.
@jkflying thinks it’s useful but slowing down too aggressive.
Land detector issues with certain types of vehicles. Proposal to make it more robust. We need to test this on more vehicles. Please
Ahmet Fehmi Ozcan has a problem with the land detector when he enables airmode. Without airmode his land detection works fine. He’ll post logs on review.px4.io
MAVSDK
No updates
MAVROS / RTPS / ROS2
No updates
Avoidance
No updates
Simulation
Thanks to @TSC21 GPS is now a sensor plugin. This allows multi-GPS stream from simulation and testing of GPS blending.
@Jaeyoung-Lim separated the airspeed sensor into its own plugin. Now multicopter is not forced to have an airspeed anymore. It’s possible to simulate multiple airspeed sensors but the HIL MAVLink message only has a field for one.
Jenkins CI consistently runs out of RA
UAVCAN
The message specification definition is still rolling. One of the main focuses at the moment is to assign subject IDs to the messages. The specifications are on a good way and it’s about filling usability holes.
Hardware call
@rroche Join the hardware call on Tuesdays (check https://www.dronecode.org/calendar/)
Hardware/Pixhawk Dev Call: Jun 16, 2020
69boards is joining Dronecode. Welcome!
Developer summit
@rroche Announcement of full schedule tomorrow. Thanks to anyone participating.
Community Q&A
No question.
In-Depth discussions
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.