1
0
mirror of https://bitbucket.org/librepilot/librepilot.git synced 2024-12-02 10:24:11 +01:00
Commit Graph

70 Commits

Author SHA1 Message Date
Laura Sebesta
8997df2438 Merge branch 'next' into kenz/diag_stack_split
Conflicts:
	flight/Modules/Stabilization/stabilization.c
2012-09-27 19:29:23 +02:00
James Cotton
4717c376c0 Stabilization: For the inner loop call the setpoint shaping version so that
the derivative term can weight the sticks and gyros independently.
2012-09-11 13:21:53 -05:00
James Cotton
11b099b1ef Stabilization: Add configuration terms for the derivative calculation to the UAVO and pass
them in to the PID system.
2012-09-11 12:29:38 -05:00
James Cotton
9763a70364 Merge remote-tracking branch 'origin/igor/pidt1' into relay_tuning
Conflicts:
	flight/Modules/Stabilization/stabilization.c
2012-09-10 03:10:26 -05:00
Igor Van Airde
8ca99739ed Changed rate controller from PID to PIDT1 to improve flight stability.
Low pass on D-Term makes the D-Term usable to improve flight stability.
Affects rate and stabilize mode.
2012-09-09 13:07:25 +02:00
Laura Sebesta
f2ffd1fb99 Changed DIAGNOSTICS flag to be individual to each module instead of global.
The upshot is the stack usage monitor can be turned on at a lower RAM threshold.
2012-08-24 12:04:05 +02:00
James Cotton
9f3c8dddd3 Add a version of the sin lookup table that is in ram instead of flash 2012-08-03 12:15:57 -05:00
James Cotton
50c7641162 Move the PID methods into a standalone library 2012-08-03 09:37:46 -05:00
James Cotton
2723ff4be3 Factor the virtual flybar code out of the stabilization module 2012-08-03 09:37:46 -05:00
James Cotton
aae0e562c6 Create a sine lookup table that is cached in flash and make relay tuning start
ot use this.
2012-08-03 09:37:46 -05:00
James Cotton
924c91ce1e Fix mistake from previous merge with duplicate the STABILIZATIONOMDE_ATTITUDE
case
2012-08-03 09:37:46 -05:00
James Cotton
8565dfbcc3 Factor the relay tuning out of the main stabilization.c file into it's own
tool.
2012-08-03 09:37:46 -05:00
James Cotton
561cf994b0 Reduce the memory footprint of the sin lookup table by using sin(x+pi) =
-sin(x).  Still just needs to move into flash and have some options about
precision when in its own library function.
2012-08-03 09:37:46 -05:00
James Cotton
d0ef95ff9a Fix idiotic bug in the stab_refactor 2012-08-03 09:37:46 -05:00
James Cotton
17878b32f3 Make autotune run within an attitude loop instead of direct rate mode. Easier
for beginners.
2012-08-03 09:37:44 -05:00
James Cotton
44e72d0a70 A big refactoring of stabilization.c to get rid of the two separate loops and
move them into one big structure.  This makes it easier to implement other
modes.
2012-08-03 09:37:44 -05:00
James Cotton
e01c5d5f87 Get online estimation of period and amplitude working 2012-08-03 09:37:44 -05:00
James Cotton
4ac8df6aa8 Make the system perform an online estimate of the period and amplitude of the
oscillation during relay tuning
2012-08-03 09:37:44 -05:00
James Cotton
17a0d3ebb4 Add a relay stabilization mode 2012-08-03 09:37:44 -05:00
James Cotton
4afd9d110e Don't zero the PIDs when the settings are updated. Fixes some quirks from
using TxPID tuning.
2012-07-03 14:19:21 +02:00
James Cotton
46a52fc5ce Apply the virtual flybar stick scaling in stabilization.c because we need to
know the true stick position to dampen the gyros appropriately.
2012-06-04 12:22:39 -05:00
James Cotton
bda3f301cb Rename VBar mode to VirtualBar 2012-06-04 12:22:39 -05:00
James Cotton
4c562b88d1 Make the amount to suppress the gyros adjustable instead of binary 2012-06-04 12:22:39 -05:00
James Cotton
59972d5570 Limit the number of degrees the vbar can deflect. This is good for fast flips
etc.
2012-06-04 12:22:39 -05:00
James Cotton
42ac4018b8 Woops. Fix really nasty bug in vbar mode I introduced. Also fix small bug in
scoping of gyro gain that made it not work.
2012-06-04 12:22:39 -05:00
James Cotton
6719ee8639 Add piro compensation option to the vbar mode 2012-06-04 12:22:39 -05:00
James Cotton
3283b99d4e For hardcode flipping allow full stick to completely override the gyro
feedback.
2012-06-04 12:22:39 -05:00
James Cotton
fe978504bc Add friendly settings for the vbar mode including the VbarTau which acts like
the time constant.
2012-06-04 12:22:39 -05:00
James Cotton
ff1b1a93cf Implement a virtual flybar in a way inspired by behavior of Align 3G which
seems apply a leaky integrator to the swash angle.  This is the similar to what
is done by Phubar (http://code.google.com/p/phubar/) as well although we refer
to the gyro term as the proportional and the flybar angle as the integral
2012-06-04 12:22:38 -05:00
Corvus Corax
64c1ac3bd6 Stabilization: Bugfix: moved queue initialisation to TaskStart where it belongs 2012-05-22 22:19:31 +02:00
Corvus Corax
d32dd0ccd4 Fixed fmod() versus fmodf() usage in flight/... 2012-03-26 04:16:54 +02:00
James Cotton
49a03a868b Merge branch 'altitudehold' into revolution_overo
Conflicts:
	flight/OpenPilot/UAVObjects.inc
	flight/Project/OpenPilotOSX/OpenPilotOSX.xcodeproj/project.pbxproj
	flight/Revolution/Makefile
	flight/Revolution/ins.c
2012-02-07 23:00:29 -06:00
James Cotton
727f0befc5 Make the gyro go a bit more regularly to stabilization but needs updating
regular CC
2012-01-25 22:50:19 -06:00
James Cotton
d2ce7761c5 Add derivative term 2012-01-03 12:37:18 -06:00
James Cotton
53cb5b67d1 Make the revo attitude now use the separate gyro object. CopterControl
attitude still needs an update.
2011-12-12 13:28:35 -06:00
James Cotton
07e3ad10af Force some floating point units to be safe 2011-11-26 15:52:25 -06:00
James Cotton
c36297227d Stabilization fix typo 2011-09-12 11:56:01 -05:00
James Cotton
b38081bb1b Stabilization: When in none mode zero the integral accumulators for rate and
attitude loops.  When not using outer loop zero that accumulator.
2011-09-11 16:33:38 -05:00
James Cotton
6945f17eba Make the StabilizationSettings.MaxRate field only apply to the stabilized
modes.  That way ManualRate can exceed MaxRate.
2011-09-11 12:27:17 -05:00
James Cotton
482bec497b Revert "Fixed modules that didn't initialize used UAVObjects correctly"
Modules should initialize their settings and the objects they source.  Not just
everything.

This reverts commit 9ad85e9b7b.
2011-08-19 09:44:38 -05:00
Corvus Corax
9ad85e9b7b Fixed modules that didn't initialize used UAVObjects correctly 2011-08-19 13:20:15 +02:00
James Cotton
bbe3c1533a Merge branch 'next' into camera_stabilization 2011-08-10 21:57:45 -05:00
James Cotton
ab288e95ab Stabilization: Small patch so that by default the integral is zeroed while
throttle is low (good for quads) but can be disabled (good for planes and also
testing helis).
2011-08-10 03:51:46 -05:00
James Cotton
96c2d24253 Merge branch 'next' into camera_stabilization
Conflicts:
	flight/CopterControl/Makefile
	flight/CopterControl/System/coptercontrol.c
	flight/Modules/Actuator/actuator.c
	flight/Modules/GPS/GPS.c
	flight/Modules/ManualControl/manualcontrol.c
	flight/Modules/Stabilization/stabilization.c
	flight/Modules/System/systemmod.c
	shared/uavobjectdefinition/manualcontrolsettings.xml
	shared/uavobjectdefinition/stabilizationdesired.xml
2011-07-30 10:06:10 +09:00
Mathieu Rondonneau
612a439199 OP-423: simplify the MODULE_INITCALL macro and remove the ordering loops 2011-07-12 20:44:32 -07:00
James Cotton
0aec3b592c Changing stabilization so that the PIDs are zeroed when not armed, but NOT when
the throttle is < 0.  This will make things like axis-lock work while armed
without throttle.  HOWEVER don't hold your stick in the arming position for a
long time or you can wind up the integrals now.
2011-07-12 14:35:32 -05:00
James Cotton
53ca934b2b Merge branch 'heli_stabilization'
Conflicts:
	flight/CopterControl/System/inc/pios_config.h
	flight/Modules/Attitude/attitude.c
	flight/Modules/ManualControl/manualcontrol.c
	flight/Modules/Stabilization/stabilization.c
2011-07-12 13:28:20 -05:00
Mathieu Rondonneau
9b9b5a1367 OP-423: remove ';' at the end of macro 2011-07-05 19:44:54 -07:00
James Cotton
ad8876c62a OP-193: Weak leveling variant on rate mode. Slowly will return to level. 2011-07-04 20:44:55 -05:00
James Cotton
a7206361d2 Slightly different way to respond to sticks in axis lock that is much more
responsive and avoids overshoot.
2011-06-26 21:59:54 -05:00