Mirror
/
ardupilot
mirror of https://github.com/ArduPilot/ardupilot
5
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
16,117 Commits 70 Branches 507 Tags 662 MiB
Commit Graph

41 Commits

Author SHA1 Message Date
Andrew Tridgell 6862791d66 PID: fixed indent-tabs-mode 2013-05-30 09:50:37 +10:00
Andrew Tridgell ba83950fc4 libraries: replace constrain() with constrain_float() 
this makes the type much more obvious. Thanks to Tobias for the
suggestion.
 2013-05-02 10:25:40 +10:00
Andrew Tridgell 6123ea2dac PID: added get_pid_4500() 
this is a version of get_pid() that returns an int16_t constrained to
-4500 to 4500. This will prevent overflow errors for large PID gains
in ArduPlane and Rover
 2013-04-01 22:17:04 +11:00
Andrew Tridgell fe964fcda0 PID: change to float input/output 
this makes the PID library a bit more flexible for smaller range
numbers. Note that this library is used on ArduPlane and Rover, not
Copter
 2013-02-09 20:36:26 +11:00
James Bielman 5631f865b2 Update floating point calculations to use floats instead of doubles. 
- Allows use of hardware floating point on the Cortex-M4.
- Added "f" suffix to floating point literals.
- Call floating point versions of stdlib math functions.
 2013-01-16 13:52:01 +11:00
Pat Hickey 39d14c8a93 PID: ported to AP_HAL 2012-12-20 14:51:31 +11:00
Andrew Tridgell 816fe9d35a PID: use NAN to flag a D reset 
this saves a byte per PID for a flag, and gives us a safe way to
reset_I() without causing a spike in D
 2012-11-27 18:41:31 +11:00
Andrew Tridgell f6fce1e238 PID: don't reset _last_error on reset_I() 
resetting _last_error when you have a non-zero D term causes the D
contribution to the next call to be massively amplified. This can
cause crazy behaviour on auto takeoff in ArduPlane if you have a
non-zero D term for the roll or picth controllers

Thanks to Chris Miser for providing the tlog that allowed this bug to
be found.
 2012-11-27 13:35:09 +11:00
Andrew Tridgell d13d117e61 APM: removed reset_I() and instead auto-reset integrator in PID library 
this prevents us resetting key integrators on waypoint change, while
still preventing old integrators being used when a PID starts to be
used again
 2012-08-28 20:18:23 +10:00
uncrustify 330c6c07e3 uncrustify libraries/PID/PID.cpp 2012-08-21 19:04:31 -07:00
Andrew Tridgell 5e8fe8d93e AP_Param: update PID libraries for new constructor syntax 2012-08-08 12:11:57 +10:00
Andrew Tridgell f6d7d1bc59 APM: changed PID library to do automatic deltat calculation 
this fixes a problem with the HDNG2RLL PID, which was using the wrong
time base and prevents similar bugs from happening in the future
 2012-07-06 19:59:18 +10:00
Andrew Tridgell 23ed5c2cee fixed imax load/save in PID 2012-02-13 16:22:53 +11:00
Andrew Tridgell 9c5a5473ab added indexes to group info structures 2012-02-13 16:22:52 +11:00
Andrew Tridgell f760b3c717 restore _fCut that was lost in AP_Param development 2012-02-13 16:22:51 +11:00
Andrew Tridgell bf4a77da81 convert libraries to new typesafe macros 2012-02-13 16:22:51 +11:00
Andrew Tridgell e5a0074312 modify the PID library for AP_Param 2012-02-13 16:22:51 +11:00
jasonshort 4ec8805216 Reverted to older PID. Restored the older filter, and removed PI command. I am no longer using PID for the quad and have switched to PIPI loops which perform much better. The D term in this class is not that great. 
git-svn-id: https://arducopter.googlecode.com/svn/trunk@3241 f9c3cf11-9bcb-44bc-f272-b75c42450872
 2011-09-04 17:47:07 +00:00
tridge60@gmail.com 5c68a04c5d PID: fixed an uninitialised variable 
we did not initialise derivative to zero

git-svn-id: https://arducopter.googlecode.com/svn/trunk@3082 f9c3cf11-9bcb-44bc-f272-b75c42450872
 2011-08-13 04:46:21 +00:00
jasonshort 915a564b8f Changed the PID derivative filter to a moving average with 4 samples. Having great luck with it and High kD when using noisy sensors. 
git-svn-id: https://arducopter.googlecode.com/svn/trunk@2957 f9c3cf11-9bcb-44bc-f272-b75c42450872
 2011-07-29 21:52:37 +00:00
jasonshort 5a22080484 I am Adding A "PI" loop to the mix. This is a need for AC2. We can optimize the internals later, but I basically duped the get_PID and removed the D term internals. 
git-svn-id: https://arducopter.googlecode.com/svn/trunk@2338 f9c3cf11-9bcb-44bc-f272-b75c42450872
 2011-05-16 16:05:49 +00:00
DrZiplok@gmail.com 944825772d Use AP_Var to store PID scaling values. 
git-svn-id: https://arducopter.googlecode.com/svn/trunk@1652 f9c3cf11-9bcb-44bc-f272-b75c42450872
 2011-02-14 04:45:31 +00:00
jasonshort c9227dbcaf fixed a scaling bug in imax 
git-svn-id: https://arducopter.googlecode.com/svn/trunk@1553 f9c3cf11-9bcb-44bc-f272-b75c42450872
 2011-01-24 02:07:25 +00:00
mich146@hotmail.com 0328f926b5 Reinstate Float loading and saving 
git-svn-id: https://arducopter.googlecode.com/svn/trunk@1507 f9c3cf11-9bcb-44bc-f272-b75c42450872
 2011-01-17 22:57:48 +00:00
jasonshort 55667fd905 removed EEPROMB references 
git-svn-id: https://arducopter.googlecode.com/svn/trunk@1315 f9c3cf11-9bcb-44bc-f272-b75c42450872
 2010-12-28 18:42:56 +00:00
james.goppert 6c10842562 Fixed PID error for low pass filter. 
git-svn-id: https://arducopter.googlecode.com/svn/trunk@1295 f9c3cf11-9bcb-44bc-f272-b75c42450872
 2010-12-27 05:59:49 +00:00
jasonshort 68eec6805a update use EEPROM class, new scaled input for RC_Channel 
git-svn-id: https://arducopter.googlecode.com/svn/trunk@1264 f9c3cf11-9bcb-44bc-f272-b75c42450872
 2010-12-26 01:17:04 +00:00
james.goppert 665dbefbd4 Added storage options to PID. 
git-svn-id: https://arducopter.googlecode.com/svn/trunk@1011 f9c3cf11-9bcb-44bc-f272-b75c42450872
 2010-12-04 10:32:19 +00:00
jasonshort 2ac2faf7a7 git-svn-id: https://arducopter.googlecode.com/svn/trunk@979 f9c3cf11-9bcb-44bc-f272-b75c42450872 2010-11-29 01:44:25 +00:00
DrZiplok@gmail.com a64c5ef782 Correct misunderstanding about the intent of the external gain array. 
git-svn-id: https://arducopter.googlecode.com/svn/trunk@978 f9c3cf11-9bcb-44bc-f272-b75c42450872
 2010-11-28 21:51:10 +00:00
DrZiplok@gmail.com 9a45c2babc Re-type delta t, explicitly size error args to get_pid. 
git-svn-id: https://arducopter.googlecode.com/svn/trunk@936 f9c3cf11-9bcb-44bc-f272-b75c42450872
 2010-11-26 04:36:43 +00:00
DrZiplok@gmail.com b0fbe39459 Remove spurious scaling 
git-svn-id: https://arducopter.googlecode.com/svn/trunk@934 f9c3cf11-9bcb-44bc-f272-b75c42450872
 2010-11-26 01:56:03 +00:00
DrZiplok@gmail.com 342b759fea Move the eeprom.h include out of the header, it's not needed there. 
git-svn-id: https://arducopter.googlecode.com/svn/trunk@933 f9c3cf11-9bcb-44bc-f272-b75c42450872
 2010-11-26 01:54:47 +00:00
DrZiplok@gmail.com 52abef58d5 More syntax, etc. fixes. Compiles OK now. 
git-svn-id: https://arducopter.googlecode.com/svn/trunk@932 f9c3cf11-9bcb-44bc-f272-b75c42450872
 2010-11-26 01:53:12 +00:00
DrZiplok@gmail.com dc40ee00ae Fix accessors, _RC 
git-svn-id: https://arducopter.googlecode.com/svn/trunk@931 f9c3cf11-9bcb-44bc-f272-b75c42450872
 2010-11-26 01:45:17 +00:00
DrZiplok 264cca6d24 With apologies to Jason for breaking his branch, clean this up and document it. 
git-svn-id: https://arducopter.googlecode.com/svn/trunk@930 f9c3cf11-9bcb-44bc-f272-b75c42450872
 2010-11-26 01:30:21 +00:00
james.goppert 3ca85f94d4 Checked for dt>0 in PID and modified ctor 
git-svn-id: https://arducopter.googlecode.com/svn/trunk@929 f9c3cf11-9bcb-44bc-f272-b75c42450872
 2010-11-26 00:06:06 +00:00
james.goppert 7598c02f51 Added load/save gain overloads for float arrays to PID, 
supports mavlink style gain get/set now


git-svn-id: https://arducopter.googlecode.com/svn/trunk@928 f9c3cf11-9bcb-44bc-f272-b75c42450872
 2010-11-25 23:26:44 +00:00
james.goppert 00e7f9b6d0 Added lowpass filter on PID lib derivative calc 
git-svn-id: https://arducopter.googlecode.com/svn/trunk@927 f9c3cf11-9bcb-44bc-f272-b75c42450872
 2010-11-25 23:14:37 +00:00
jasonshort 391e4268c4 git-svn-id: https://arducopter.googlecode.com/svn/trunk@859 f9c3cf11-9bcb-44bc-f272-b75c42450872 2010-11-19 07:18:40 +00:00
jasonshort e7b26da27d A simple OO PID library 
git-svn-id: https://arducopter.googlecode.com/svn/trunk@742 f9c3cf11-9bcb-44bc-f272-b75c42450872
 2010-10-28 04:59:24 +00:00