[WIP]: mc_att_control split out rate controller and sensor aggregation into new modules #12225

dagar · 2019-06-10T00:54:27Z

~~Requires #12207 before merging.~~
Background - #12207

The multicopter attitude controller currently contains a rate controller, attitude controller, and logic for handling multiple gyros.

mc_att_control components

polls primary sensor_gyro based on current sensor_selection, then adds corrections and in run bias
rate controller
attitude controller

This PR splits mc_att_control into 3 separate modules, each running out of a WQ. We can afford to do this because each will be running out of an appropriate WQ thread, rather than as a new standalone task.

sensors/vehicle_angular_velocity

subscribes to all gyros directly (sensor_gyro), estimated bias (sensor_bias), primary gyro from sensors (sensor_selection), and optional corrections (sensor_corrections)
publishes a single vehicle_angular_velocity

mc_rate_control

simpler rate controller that runs on new vehicle_angular_velocity updates
publishes actuator_controls_0 and rate_ctrl_status

mc_att_control

simpler attitude controller than runs on new vehicle_attitude updates
publishes vehicle_rate_setpoint

dagar · 2019-06-10T00:58:30Z

The overall diff is actually quite small. The majority of the change in the main loop is one level of indentation.

Junkim3DR · 2019-06-18T17:29:50Z

Tested on Pixhawk 4mini v5:

Modes Tested

Position Mode: Good.
Altitude Mode: Good.
Stabilized Mode: Good.
Mission Plan Mode (Automated): Good.
RTL (Automated): Good.

Procedure
Arm and Take off in position mode, after flying for approximately one minute, switched to altitude then stabilized mode proceed to switch to mission plan mode then make sure that vehicle follows all waypoints as shown in QGC, once completed all waypoint activate RTL and see landing behaviour.

Notes:
No issues noted, good flight in general.

Log:
PR 12225

https://review.px4.io/plot_app?log=8d631448-e7fa-4e28-8f04-726988fb2211

Master

https://review.px4.io/plot_app?log=86a79be0-549e-4d83-a8af-ad2a5f2b6c88

Junkim3DR · 2019-06-18T18:14:00Z

Tested on Pixhawk 2 Cube v3:

Modes Tested

Position Mode: Good.
Altitude Mode: Good.
Stabilized Mode: Good.
Mission Plan Mode (Automated): Good.
RTL (Automated): Good.

Procedure
Arm and Take off in position mode, after flying for approximately one minute, switched to altitude then stabilized mode proceed to switch to mission plan mode then make sure that vehicle follows all waypoints as shown in QGC, once completed all waypoint activate RTL and see landing behaviour.

Notes:
No issues noted, good flight in general.

Log:
PR 12225

Master

jorge789 · 2019-06-18T21:43:27Z

Tested on PixRacer V4:

Modes Tested

Position Mode: Good.
Altitude Mode: Good.
Stabilized Mode: Good.
Mission Plan Mode (Automated): Good.
RTL (Automated): Good.
Procedure
Arm and Take off in position mode, after flying for approximately one minute, switched to altitude then stabilized mode proceed to switch to mission plan mode then make sure that vehicle follows all waypoints as shown in QGC, once completed all waypoint activate RTL and see landing behaviour.

Notes:
No issues noted, good flight in general.

Log:
PR 12225
https://review.px4.io/plot_app?log=f2af7211-1b9a-4f8c-9193-29bbba244fc9

https://review.px4.io/plot_app?log=5d6fcd40-dc85-4bf2-a042-9dc7788042b4

Master:
https://review.px4.io/plot_app?log=d8565407-cab5-43f6-9abc-5cfcea1b4fcf

MaEtUgR

Looks very good, I had some small questions but I generally like the direction this takes.
It all seems fine in SITL, I don't see any issues.

MaEtUgR · 2019-06-24T15:27:44Z

msg/vehicle_angular_velocity.msg

+
+uint64 timestamp_sample		# the timestamp of the raw data (microseconds)
+
+float32 rollspeed		# Bias corrected angular velocity about X body axis in rad/s


Could we make that an array?

MaEtUgR · 2019-06-24T16:55:45Z

src/modules/sensors/vehicle_angular_velocity/VehicleAngularVelocity.hpp

+ * POSSIBILITY OF SUCH DAMAGE.
+ *
+ ****************************************************************************/
+


I think a very brief description of the class would help reading through the code.

src/modules/mc_att_control/mc_att_control_main.cpp

MaEtUgR · 2019-06-24T16:59:25Z

msg/CMakeLists.txt

@@ -123,6 +123,7 @@ set(msg_files
 	ulog_stream.msg
 	ulog_stream_ack.msg
 	vehicle_air_data.msg
+	vehicle_angular_velocity.msg


What about vehicle_angular_rate? I know this term is used a lot in this pr, I'm just suggesting it because terminoligy would be consistent with "rate controller". What do you think?

I don't feel that strongly about it as long is we try to be consistent at some level (with literature, ROS, internally, etc). When I first sketched this out I was just calling it vehicle body rate or even vehicle rates.

MaEtUgR · 2019-06-24T17:21:57Z

src/modules/mc_att_control/mc_att_control_main.cpp

+bool
+MulticopterAttitudeControl::init()
+{
+	return selected_gyro_update();


Do you select the gyro on module start "only"?

Why was I already outdated? 🤷‍♂

I think it was commenting on something in the 1st commit (mc_att_control changes) that changed again in the 2nd (moving this out of mc_att_control entirely)?

Do you select the gyro on module start "only"?

It also updates if sensor_selection (from sensors voting) changes as well.

src/modules/mc_att_control/mc_att_control_main.cpp

dagar · 2019-08-06T18:26:43Z

Continued in #12650.

dagar added the Admin: Enhancement (improvement) 💡 label Jun 10, 2019

dagar mentioned this pull request Jun 10, 2019

uORB Subscription callbacks with WorkItem scheduling on publication #12207

Merged

dagar added the Multirotor 🛸 label Jun 10, 2019

dagar force-pushed the pr-orb_subscription_callback-mc_att_control branch 2 times, most recently from 8a5e1b5 to c5f9664 Compare June 13, 2019 03:00

dagar changed the base branch from pr-orb_subscription_callback to master June 13, 2019 03:00

dagar force-pushed the pr-orb_subscription_callback-mc_att_control branch 2 times, most recently from 6516ae8 to 479756b Compare June 13, 2019 15:02

dagar changed the base branch from master to pr-orb_subscription_callback June 13, 2019 15:02

dagar force-pushed the pr-orb_subscription_callback branch 2 times, most recently from 312999b to a67e42b Compare June 13, 2019 18:07

dagar force-pushed the pr-orb_subscription_callback-mc_att_control branch 2 times, most recently from b7c96c9 to ca93abd Compare June 13, 2019 18:38

weekly-digest bot mentioned this pull request Jun 16, 2019

Weekly Digest (9 June, 2019 - 16 June, 2019) #12290

Closed

dagar changed the title ~~[WIP]: mc_att_control move to WQ with uORB callback scheduling~~ mc_att_control move to WQ with uORB callback scheduling Jun 17, 2019

dagar force-pushed the pr-orb_subscription_callback-mc_att_control branch from ca93abd to 3ff00e0 Compare June 17, 2019 20:17

dagar requested review from a team and MaEtUgR June 17, 2019 20:17

dagar marked this pull request as ready for review June 17, 2019 20:17

dagar changed the base branch from pr-orb_subscription_callback to master June 17, 2019 20:17

dagar force-pushed the pr-orb_subscription_callback-mc_att_control branch from 3ff00e0 to 5e4cca9 Compare June 18, 2019 00:11

weekly-digest bot mentioned this pull request Jun 23, 2019

Weekly Digest (16 June, 2019 - 23 June, 2019) #12318

Closed

MaEtUgR reviewed Jun 24, 2019

View reviewed changes

dagar force-pushed the pr-orb_subscription_callback-mc_att_control branch 3 times, most recently from 0e4b030 to b045539 Compare June 27, 2019 05:28

dagar force-pushed the pr-orb_subscription_callback-mc_att_control branch 3 times, most recently from 444c22f to cf50d84 Compare June 28, 2019 15:27

weekly-digest bot mentioned this pull request Jun 30, 2019

Weekly Digest (23 June, 2019 - 30 June, 2019) #12374

Closed

dagar mentioned this pull request Jul 1, 2019

Rate setpoint dropout during transition #12359

Closed

dagar changed the title ~~mc_att_control move to WQ with uORB callback scheduling~~ [WIP]: mc_att_control split out rate controller and sensor aggregation into new modules Jul 1, 2019

weekly-digest bot mentioned this pull request Jul 7, 2019

Weekly Digest (30 June, 2019 - 7 July, 2019) #12430

Closed

dagar mentioned this pull request Jul 13, 2019

Remove Euler angle rate names #12451

Closed

dagar mentioned this pull request Jul 26, 2019

px4fmu move to WQ with uORB callback scheduling #12224

Merged

7 tasks

dagar added 3 commits July 26, 2019 21:21

mc_att_control move to WQ with uORB callback scheduling

1cc5337

mc_att_control split out sensor_gyro aggregation

13634dd

mc_rate_control split out of mc_att_control

d127813

dagar force-pushed the pr-orb_subscription_callback-mc_att_control branch from cf50d84 to d127813 Compare July 27, 2019 01:31

weekly-digest bot mentioned this pull request Jul 28, 2019

Weekly Digest (21 July, 2019 - 28 July, 2019) #12569

Closed

dagar mentioned this pull request Aug 1, 2019

sensors: create vehicle_angular_velocity module #12596

Merged

dagar closed this Aug 6, 2019

MaEtUgR mentioned this pull request Oct 13, 2019

mc_att_control: move rate control to RateControl class #13173

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

[WIP]: mc_att_control split out rate controller and sensor aggregation into new modules #12225

[WIP]: mc_att_control split out rate controller and sensor aggregation into new modules #12225

dagar commented Jun 10, 2019 •

edited

Loading

dagar commented Jun 10, 2019

Junkim3DR commented Jun 18, 2019 •

edited

Loading

Junkim3DR commented Jun 18, 2019

jorge789 commented Jun 18, 2019

MaEtUgR left a comment •

edited

Loading

MaEtUgR Jun 24, 2019

MaEtUgR Jun 24, 2019

MaEtUgR Jun 24, 2019 •

edited

Loading

dagar Jun 24, 2019

MaEtUgR Jun 24, 2019

MaEtUgR Jun 24, 2019

dagar Jun 24, 2019

dagar Jun 28, 2019

dagar commented Aug 6, 2019


		uint64 timestamp_sample # the timestamp of the raw data (microseconds)

		float32 rollspeed # Bias corrected angular velocity about X body axis in rad/s

[WIP]: mc_att_control split out rate controller and sensor aggregation into new modules #12225

[WIP]: mc_att_control split out rate controller and sensor aggregation into new modules #12225

Conversation

dagar commented Jun 10, 2019 • edited Loading

sensors/vehicle_angular_velocity

mc_rate_control

mc_att_control

dagar commented Jun 10, 2019

Junkim3DR commented Jun 18, 2019 • edited Loading

Tested on Pixhawk 4mini v5:

Junkim3DR commented Jun 18, 2019

Tested on Pixhawk 2 Cube v3:

jorge789 commented Jun 18, 2019

Tested on PixRacer V4:

MaEtUgR left a comment • edited Loading

Choose a reason for hiding this comment

MaEtUgR Jun 24, 2019

Choose a reason for hiding this comment

MaEtUgR Jun 24, 2019

Choose a reason for hiding this comment

MaEtUgR Jun 24, 2019 • edited Loading

Choose a reason for hiding this comment

dagar Jun 24, 2019

Choose a reason for hiding this comment

MaEtUgR Jun 24, 2019

Choose a reason for hiding this comment

MaEtUgR Jun 24, 2019

Choose a reason for hiding this comment

dagar Jun 24, 2019

Choose a reason for hiding this comment

dagar Jun 28, 2019

Choose a reason for hiding this comment

dagar commented Aug 6, 2019

dagar commented Jun 10, 2019 •

edited

Loading

Junkim3DR commented Jun 18, 2019 •

edited

Loading

MaEtUgR left a comment •

edited

Loading

MaEtUgR Jun 24, 2019 •

edited

Loading