6f1cee6406
AP_AHRS: fixed position estimate with 2D fix or low satellite count
...
we should still use the GPS for position fixes when we have a low
number of satellites, but we should stop using it for velocity
estimates and attitude correction.
2013-05-07 10:52:14 +10:00
ceda7d26b0
AHRS: add correct_centrifugal and flags structure
2013-05-06 15:32:06 +09:00
38fc0e61c6
AP_AHRS: optimise yaw drift correction
...
use new vector2 cross product, and factor out the z component of a dcm
mul_transpose() to reduce the number of floating point operations for
a yaw drift correction cycle
2013-05-05 13:51:45 +10:00
ee2daf25b6
AHRS: perf improvement by caching declination vector
2013-05-05 12:58:43 +10:00
943a1d8c8d
AP_AHRS: added AHRS_GPS_MINSATS option
...
if the number of visible satellites is below AHRS_GPS_MINSATS then
don't use the GPS for acceleration correction for attitude
2013-05-05 12:48:01 +10:00
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
f5d6fb6216
AP_AHRS: added _kp gain reduction for catapult launch
...
this should reduce the tendency of aircraft to nose down hard on
initial launch at high g
2013-04-28 14:47:59 +10:00
fd2cd0d1c2
DCM: fixed formatting
2013-04-23 08:57:44 +10:00
610069e729
DCM: fixed airspeed bug
...
we incorrectly went via the wind estimate when we had a working airspeed sensor
2013-04-23 08:57:44 +10:00
a4d25f5a82
AHRS: removed AHRS_BARO_USE option
...
this option has caused users too much trouble. The vertical velocity
is too noisy from the baro
2013-04-15 10:52:31 +10:00
8459da202c
AP_AHRS: prevents compass flyaways for plane and rover
...
this switches to the GPS for yaw if the compass has dragged us off by
more than 45 degrees from the GPS heading, and the wind speed is less
than 80% of the ground speed.
2013-03-29 13:48:25 +11:00
24044dc0c4
AHRS: add support for GPS fix type 2D
2013-03-27 11:41:43 +09:00
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
3ac3aeb1b1
AHRS: fixed build on ARM
2013-01-02 14:45:09 +11:00
1c08f176ea
AP_AHRS: add get_accel_ef method to return earth frame accelerometer values for use in ArduCopter's inertial nav and accel based throttle
2013-01-02 09:11:25 +11:00
ceb3f577d8
libraries: use new math functions
2012-12-20 14:53:22 +11:00
8b6ad6d559
SITL: removed SITL_debug()
...
not needed any more
2012-12-20 14:51:36 +11:00
5840ded767
AHRS: removed constrain() defines
2012-12-20 14:51:34 +11:00
9bf69d4e0d
AP_AHRS: port to AP_HAL
2012-12-20 14:51:29 +11:00
8561f08b2f
AP_AHRS: prevent a large delta_t from affecting DCM solution
...
this should fix large changes in attitude directly after arming
ArduCopter
2012-11-13 15:33:10 +11:00
83ae8e47be
AP_AHRS: replace IMU with INS and add roll and pitch trim
2012-11-07 19:20:32 +09:00
fe47990dab
AP_AHRS: removed the 6 sats min
...
this would put us into dead-reckoning mode
2012-11-06 14:26:49 +11:00
6a24bdec05
AP_AHRS: require at least 6 satellites to use the GPS for velocity
...
logs of a recent flight show the velocity estimate can be very poor if
the GPS can see 5 satellites or less
2012-11-05 20:07:37 +11:00
ea40432235
AP_AHRS: use GPS vertical velocity when available
2012-11-05 14:58:40 +11:00
00bf548f87
AHRS: limit wind speed estimate changes
...
sudden yaw changes can cause bad wind estimates
2012-10-11 11:51:24 +11:00
ccf8ba3ce9
DCM: limit the contribution of the baro to vertical acceleration
...
limit to a max of 0.5g
2012-09-20 17:34:32 +10:00
fba1692d69
AHRS: added AHRS_BARO_USE parameter
...
allow disabling of the use of the barometer for vertical acceleration
compensation
2012-09-12 14:42:04 +10:00
de28cc8b28
AHRS: added AHRS_WIND_MAX option
...
this allows APM to cope better with airspeed sensor failure, but
ensuring airspeed stays within AHRS_WIND_MAX of ground speed
2012-09-08 11:37:07 +10:00
2be785899b
AHRS: added airspeed_estimate() function
...
this allows the APM code to use an airspeed estimate for navigation
2012-08-25 17:49:26 +10:00
31a4442904
uncrustify libraries/AP_AHRS/AP_AHRS_DCM.cpp
2012-08-21 19:19:51 -07:00
134cd51d17
AHRS: moved var_info[] into top level AP_AHRS class
2012-08-21 15:58:09 +10:00
ac72db8f00
AHRS: added set_fast_gains() method
...
this will be used by arducopter on the ground
2012-08-21 15:38:06 +10:00
14cdbd36cb
AHRS: check for bad values in the error before they can affect DCM
...
this should fix the DCM renorm errors in autotest, probably caused by
bad climb rates
2012-08-18 13:54:11 +10:00
d9cc967508
AHRS: force an extra read of the compass on startup
...
the first read from the compass can be bad. This ensures we have a
good value when getting the initial AHRS yaw.
Thanks to Randy and Jason for the bug report!
2012-08-17 17:57:39 +10:00
19a5d11669
AHRS: use airspeed for wind in forward flight
...
when not turning we can use the airspeed sensor directly to calculate
the wind speed.
2012-08-12 10:54:25 +10:00
f5b7ae213c
AHRS: added AHRS_GPS_USE parameter
...
this allows for the GPS to be disables for position updates, which
allows for testing dead-reckoning
2012-08-11 22:06:03 +10:00
779dacfd24
AHRS: added wind estimation code
...
this allows us to estimate the wind while we have GPS lock, then use
that estimate in the long term dead reckoning
2012-08-11 22:06:03 +10:00
af4071894e
AHRS: added long-term dead-reckoning
...
this uses airspeed (if available) or last GPS ground speed to update
our position estimate in AHRS
2012-08-11 12:00:31 +10:00
898a5af692
AP_Param: update AHRS for new constructor syntax
2012-08-08 12:11:57 +10:00
468170b589
AP_AHRS_DCM: moved var_info to belong to AP_AHRS_DCM instead of AP_AHRS
2012-07-28 14:16:15 +09:00
9cd98e79dd
AHRS: normalise GA_b before computing error vector only if too large
...
When GA_b is small the direction of the vector is unreliable, so
normalising can exacerbate the error in the direction
2012-07-23 17:16:50 +10:00
5f5ab9c07c
AHRS: if we don't have gps correction enabled, zero the yaw correction
2012-07-20 17:21:04 +10:00
c937738732
AHRS: allow the gps/accelerometers to control yaw when rolled
...
this solves a problem with yaw estimation when at high roll angles,
when the compass code becomes ineffective
2012-07-20 11:50:04 +10:00
a469682b7b
fix param config names
2012-07-14 15:34:15 +08:00
0f9fb2c64b
AHRS: added AHRS_GPS_GAIN parameter
...
this allows control of how much GPS velocity information is used in
doing centripetal correction in DCM
2012-07-10 16:22:48 +10:00
302696951a
AHRS: make the yaw independent drift correction optional and disable
...
the new correction algorithm copes poorly with gyro drift, leading to
signification attitude errors in the face of drift
2012-07-06 15:11:30 +10:00
8c5ef3e60f
AHRS: only enable barometer for AHRS if it is smooth enough
...
if we used less than 5 samples to compute the pressure then don't use
it for climb rate
2012-07-05 16:29:21 +10:00
2cd6da2539
AHRS: only use GPS for yaw when compass is not being used
...
this avoids having two competing controllers
2012-07-05 13:00:46 +10:00
adfa97b6f8
AHRS: enable barometer for vertical velocity
...
this re-enables the barometer for vertical velocity information
for drift correction, now that we have a better filter
on the climb rate
2012-07-05 13:00:46 +10:00
3849ca8b5c
AHRS: make DCM drift correction not rely on accurate yaw
...
this uses a new formulation of the GPS based drift correction from
Bill Premerlani that rotates the error vector to avoid relying on
accurate yaw. This means we should get accurate roll/pitch correction
even with lots of magnetometer interference
It also makes it possible to fly a multicopter with no compass. It can
even navigate and correct yaw (slowly!)
2012-07-04 17:00:40 +10:00
1314e4f872
AHRS: make P gain on PI roll/pitch controller tunable
...
this is to make it easier for Craig and Alan to tune
2012-07-04 12:42:46 +10:00
16c55c64d4
AHRS: don't add the P terms in _omega
...
this can lead feedback via the _P_gain()
2012-06-29 15:10:51 +10:00
37fc6c46b9
AHRS: disable barometer for vertical acceleration
...
this seems to have been the cause of the 'flips' seen by Marco and
others. Testing by Craig and Alan shows that the flips are gone when
the barometric acceleration is removed.
It looks like a 5 point average filter is not enough to keep the
vertical acceleteration noise low. With high noise in the z axes, the
x and y axes are scaled back when the ge vector is normalised.
2012-06-28 11:12:49 +10:00
58c4d458dc
AHRS: re-instate new DCM drift correction code
...
This reverts commit 078489638d47fbaffde7c51249e36b5a8fc4ef9d.
2012-06-28 11:09:22 +10:00
fd241692d0
AHRS: revert to the old drift correction algorithm
...
we need to work out why Craigs quad flipped today before we enable
this new drift correction
2012-06-27 18:16:41 +10:00
8e044a560b
AHRS: include P term in omega
...
thanks to Jon for suggesting this
2012-06-27 16:01:50 +10:00
37e2f2ee2c
AHRS: implement spin rate limits
...
this follows the method that Bill developed in his fastRotations
paper. We've demonstrated that this is indeed needed in APM, as we
were able to produce the 'dizzy' effects in both the ArduPlane and
ArduCopter simulator
2012-06-27 16:01:50 +10:00
e4a245179c
AHRS: cope with copters with no compass
...
if a copter doesn't have a compass, we can't use the GPS for gyro
drift correction
2012-06-27 16:01:50 +10:00
3a41ad8e7c
AHRS: normalize the ge vector in drift correction, and use barometer
...
The normalisation ensures the error term scales uniformly with
different accelerations.
The barometer is used for vertical acceleration estimation
2012-06-27 16:01:50 +10:00
d230690b7b
AHRS: brought DCM more inline with Bill's implementation
...
omega_I applied continuously. _ki larger. Stop integrating when _omega.length()>20
The key change was the scaling of ge to ensure the error is not
quadratic
2012-06-27 16:01:49 +10:00
d4c8041c51
AHRS: first successful version of Bills new drift correction system
...
This makes 3 major changes:
1) fixes the scaling of the yaw drift correction term to fix the time
constant
2) don't integrate the mag vector over multiple readings
3) accumulate omega_I changes over 15 seconds before applying, to try
to prevent omega_I picking up short term responses
2012-06-27 16:01:49 +10:00
0f5c22bd8e
AHRS: implement Bills new drift correction algorithm
...
this is an initial implementation of this paper:
http://gentlenav.googlecode.com/files/RollPitchDriftCompensation.pdf
2012-06-27 16:01:49 +10:00
f274df454a
DCM: fixed a bug when using GPS for yaw correction
...
When using GPS for yaw correction we need to apply the x and y omegaI
corrections from the _omega_I_sum in the period before we get to the
minimum ground speed for GPS yaw correction. Otherwise we get a large
sudden omega_I change on takeoff.
2012-05-21 12:15:06 +10:00
3309dec205
AP_AHRS_DCM.cpp: Added comments that follow the new parse structure
2012-04-25 12:00:42 -05:00
2a3af369b6
DCM: buffer omega_I changes over 10 seconds
...
this buffers up _omega_I changes in _omega_I_sum over a period of 10
seconds, applying the slope limit only when _omega_I_sum is
transferred to _omega_I.
The result is a huge improvement in the ability of _omega_I to track
gyro drift over the long term.
2012-04-23 12:13:25 +10:00
f27d85cb8d
DCM: drop the 'drop z' method
...
the 'drop z' method reduced the impact of noise on omegaI, but it also
made us more sensitive to errors in accelerometer calibration and
scaling, as demonstated by the logs from Gabor here:
http://diydrones.com/xn/detail/705844:Comment:834373
Simulation testing shows that the other noise suppression methods
applied in the DCM code, in particular the slope limiting on omegaI
the removal of the weighting and the upcoming use of a _omega_I_sum
buffer have reduced the impact of noise enough that we can now safely
include z in the acceleration calculation.
2012-04-23 12:13:24 +10:00
8e90aeea4e
AHRS: added AHRS_YAW_P parameter
...
this allows users to change the yaw gain in DCM
2012-04-16 20:55:13 +10:00
b549b88e5e
AHRS: fixed error_yaw reporting with 2 MAVLink connections
...
when a user first connects with USB, and later switches to the
telemetry port without restarting we were getting zero for error_yaw
in the logs, as AHRS.get_error_yaw() was being called twice.
This ensures we give the last value after the counter is reset
2012-03-29 12:39:53 +11:00
4f82134a5e
DCM: use the new rotate() method from AP_Math
...
this allows us to use a tested and optimised rotation method
2012-03-19 17:29:02 +11:00
d4bb068d5b
AHRS: adapt the DCM library to the AHRS framework
2012-03-19 17:29:02 +11:00
9d75ffeba1
AHRS: rename DCM and Quaternion implementions ready for AHRS class
...
this is the first step to creating a general AHRS class for ArduPilot
2012-03-19 17:29:01 +11:00
Andrew Tridgell
Randy Mackay
jschall
James Bielman
Pat Hickey
uncrustify
Michael Oborne
Adam M Rivera