- ekf2 can now run in multi-instance mode (currently up to 9 instances)
- in multi mode all estimates are published to alternate topics (eg estimator_attitude instead of vehicle_attitude)
- new ekf2 selector runs in multi-instance mode to monitor and compare all instances, selecting a primary (eg N x estimator_attitude => vehicle_attitude)
- sensors module accel & gyro inconsistency checks are now relative to the mean of all instances, rather than the current primary (when active ekf2 selector is responsible for choosing primary accel & gyro)
- existing consumers of estimator_status must check estimator_selector_status to select current primary instance status
- ekf2 single instance mode is still fully supported and the default
Co-authored-by: Paul Riseborough <gncsolns@gmail.com>
- add new parameter `IMU_GYRO_FFT_EN` to start
- add 75% overlap in buffer to increase FFT update rate
- space out FFT calls (no more than 1 per cycle)
- increase `IMU_GYRO_FFT_MIN` default
- decrease main stack usage
- drivers/tone_alarm: move to ModuleBase and purge CDev (/dev/tone_alarm0)
- drivers/tone_alarm: only run on tune_control publication (or scheduled note) rather than continuously
- drivers/tone_alarm: use HRT to schedule tone stop (prevents potential disruption)
- msg/tune_control: add tune_id numbering
- systemcmds/tune_control: add "error" special case tune_id
- move all tune_control publication to new uORB::PublicationQueued<>
- start tone_alarm immediately after board defaults are loaded to fix potential startup issues
- for SITL (or other boards with no TONE output) print common messages (startup, error, etc)
Nuttx now supports sh and source (.)
sh will open a new process for each invocation.
This means the child can not modify the parent
env. So we must use . to matain how nuttx worked.
Since rc.vehicle_setup is used in both we use
source and alis as we did with sh.
* [WIP] i2c_spi_buses: add '-q' for quiet startup flag
And enable for optional board sensors.
* ROMFS: rc.sensors try starting all optional I2C sensors quietly
Co-authored-by: Daniel Agar <daniel@agar.ca>
- in practice this is mostly useful for identifying incorrect rotations
which we mostly have in 45 degree increments
- handling a vehicle on the ground can easily disturb one mag by more than 30 degrees, so this is often distracting noise
- the values of the parameters GND_MAX_ANG and GND_WHEEL_BASE are outdated. They belonged to another Rover setup.
- in the mixer file rover_diff_and_servo.main.mix the steering was controlled by roll, but in PX4 rover steering is controlled by yaw. And this was the reason why the attitude control did not work correctly
- the DF Robot GPX:Asurada rover has actual a steering angle of 60 degrees. And and wheel base of 0.17m. Parameter values in the airframe file are changed to this values
- avoids the need for ekf2_timestamp publications by q and lpe
- adds logger to the lockstep cycle and makes it poll on ekf2_timestamps
or vehicle_attitude. This avoids dropped samples (required for replay).
Before #14212 the velocity control gains used in the multicopter
position controller were defined as a scale between velocity error in
one axis (or it's integral and derivative respectively) and the unit
thrust vector. The problem with this is that the normalization of the
unit thrust vector changes per vehicle or even vehicle configuration
as 0 and 100% thrust get a different physical response. That's why
the gains are now defined as scale between velocity error
(integral/derivative) and the output acceleration in m/s².
Only have it higher for VTOL and fixed wing.
Multicopter position controlled flight is in our experience always <1g.
Acrobatic flying definitely exceeds the acceleration but if control
doesn't rely on the GPS velocity and position there shouldn't be any
problem.
@fury1895 reported very valuable feedback from testing
the acceleration feed-forward on VTOL:
> MPC_JERK_MAX 4.5 (from 5 on it felt too aggressive)
> MPC_JERK_AUTO 4
> some hovering, some transitions, and a mission. Everything good.
> I'd say you feel the difference in position mode and you see it in
> Auto modes. Great improvement!
Daniel Agar
Beat Küng
David Sidrane
jaeyoung
Matthias Grob
Katrin Moritz
Daniel Leonard Robinson
Katrin
xdwgood
Stone White
斯东Stone
baumanta
Hamish Willee
flbernier
Nico van Duijn
AlexKlimaj
Julian Oes