tune
Syntax
Description
adjusts the properties of the tunedMeasureNoise
= tune(filter
,measureNoise
,sensorData
,groundTruth
)insfilterErrorState
filter object,
filter
, and measurement noises to reduce the root-mean-squared (RMS)
state estimation error between the fused sensor data and the ground truth. The function also
returns the tuned measurement noise, tunedMeasureNoise
. The function
uses the property values in the filter and the measurement noise provided in the
measureNoise
structure as the initial estimate for the optimization
algorithm.
specifies the tuning configuration based on a tunedMeasureNoise
= tune(___,config
)tunerconfig
object,
config
.
Examples
Tune insfilterErrorState
to Optimize Pose Estimate
Load the recorded sensor data and ground truth data.
load('insfilterErrorStateTuneData.mat');
Create tables for the sensor data and the truth data.
sensorData = table(Accelerometer,Gyroscope, ... GPSPosition,GPSVelocity,MVOOrientation, ... MVOPosition); groundTruth = table(Orientation,Position);
Create an insfilterErrorState
filter object.
filter = insfilterErrorState('State',initialState, ... 'StateCovariance',initialStateCovariance);
Create a tuner configuration object for the filter. Use the tuner noise function to obtain a set of initial sensor noises used in the filter.
cfg = tunerconfig('insfilterErrorState','MaxIterations',40); measNoise = tunernoise('insfilterErrorState')
measNoise = struct with fields:
MVOOrientationNoise: 1
MVOPositionNoise: 1
GPSPositionNoise: 1
GPSVelocityNoise: 1
Tune the filter and obtain the tuned parameters.
tunedmn = tune(filter,measNoise,sensorData, ...
groundTruth,cfg);
Iteration Parameter Metric _________ _________ ______ 1 AccelerometerNoise 4.1291 1 GyroscopeNoise 4.1291 1 AccelerometerBiasNoise 4.1290 1 GyroscopeBiasNoise 4.1290 1 GPSPositionNoise 4.0213 1 GPSVelocityNoise 4.0051 1 MVOPositionNoise 3.9949 1 MVOOrientationNoise 3.9886 2 AccelerometerNoise 3.9886 2 GyroscopeNoise 3.9886 2 AccelerometerBiasNoise 3.9886 2 GyroscopeBiasNoise 3.9886 2 GPSPositionNoise 3.8381 2 GPSVelocityNoise 3.8268 2 MVOPositionNoise 3.8219 2 MVOOrientationNoise 3.8035 3 AccelerometerNoise 3.8035 3 GyroscopeNoise 3.8035 3 AccelerometerBiasNoise 3.8035 3 GyroscopeBiasNoise 3.8035 3 GPSPositionNoise 3.6299 3 GPSVelocityNoise 3.6276 3 MVOPositionNoise 3.6241 3 MVOOrientationNoise 3.5911 4 AccelerometerNoise 3.5911 4 GyroscopeNoise 3.5911 4 AccelerometerBiasNoise 3.5911 4 GyroscopeBiasNoise 3.5911 4 GPSPositionNoise 3.1728 4 GPSVelocityNoise 3.1401 4 MVOPositionNoise 2.7686 4 MVOOrientationNoise 2.6632 5 AccelerometerNoise 2.6632 5 GyroscopeNoise 2.6632 5 AccelerometerBiasNoise 2.6632 5 GyroscopeBiasNoise 2.6632 5 GPSPositionNoise 2.3242 5 GPSVelocityNoise 2.2291 5 MVOPositionNoise 2.2291 5 MVOOrientationNoise 2.0904 6 AccelerometerNoise 2.0903 6 GyroscopeNoise 2.0903 6 AccelerometerBiasNoise 2.0903 6 GyroscopeBiasNoise 2.0903 6 GPSPositionNoise 2.0903 6 GPSVelocityNoise 2.0141 6 MVOPositionNoise 1.9952 6 MVOOrientationNoise 1.8497 7 AccelerometerNoise 1.8497 7 GyroscopeNoise 1.8496 7 AccelerometerBiasNoise 1.8496 7 GyroscopeBiasNoise 1.8496 7 GPSPositionNoise 1.8398 7 GPSVelocityNoise 1.7528 7 MVOPositionNoise 1.7362 7 MVOOrientationNoise 1.5762 8 AccelerometerNoise 1.5762 8 GyroscopeNoise 1.5762 8 AccelerometerBiasNoise 1.5762 8 GyroscopeBiasNoise 1.5762 8 GPSPositionNoise 1.5762 8 GPSVelocityNoise 1.5107 8 MVOPositionNoise 1.4786 8 MVOOrientationNoise 1.3308 9 AccelerometerNoise 1.3308 9 GyroscopeNoise 1.3308 9 AccelerometerBiasNoise 1.3308 9 GyroscopeBiasNoise 1.3308 9 GPSPositionNoise 1.3308 9 GPSVelocityNoise 1.2934 9 MVOPositionNoise 1.2525 9 MVOOrientationNoise 1.1462 10 AccelerometerNoise 1.1462 10 GyroscopeNoise 1.1462 10 AccelerometerBiasNoise 1.1462 10 GyroscopeBiasNoise 1.1462 10 GPSPositionNoise 1.1443 10 GPSVelocityNoise 1.1332 10 MVOPositionNoise 1.0964 10 MVOOrientationNoise 1.0382 11 AccelerometerNoise 1.0382 11 GyroscopeNoise 1.0382 11 AccelerometerBiasNoise 1.0382 11 GyroscopeBiasNoise 1.0382 11 GPSPositionNoise 1.0348 11 GPSVelocityNoise 1.0348 11 MVOPositionNoise 1.0081 11 MVOOrientationNoise 0.9734 12 AccelerometerNoise 0.9734 12 GyroscopeNoise 0.9734 12 AccelerometerBiasNoise 0.9734 12 GyroscopeBiasNoise 0.9734 12 GPSPositionNoise 0.9693 12 GPSVelocityNoise 0.9682 12 MVOPositionNoise 0.9488 12 MVOOrientationNoise 0.9244 13 AccelerometerNoise 0.9244 13 GyroscopeNoise 0.9244 13 AccelerometerBiasNoise 0.9244 13 GyroscopeBiasNoise 0.9244 13 GPSPositionNoise 0.9203 13 GPSVelocityNoise 0.9199 13 MVOPositionNoise 0.9045 13 MVOOrientationNoise 0.8846 14 AccelerometerNoise 0.8846 14 GyroscopeNoise 0.8846 14 AccelerometerBiasNoise 0.8845 14 GyroscopeBiasNoise 0.8845 14 GPSPositionNoise 0.8807 14 GPSVelocityNoise 0.8807 14 MVOPositionNoise 0.8659 14 MVOOrientationNoise 0.8501 15 AccelerometerNoise 0.8501 15 GyroscopeNoise 0.8501 15 AccelerometerBiasNoise 0.8500 15 GyroscopeBiasNoise 0.8500 15 GPSPositionNoise 0.8457 15 GPSVelocityNoise 0.8453 15 MVOPositionNoise 0.8299 15 MVOOrientationNoise 0.8173 16 AccelerometerNoise 0.8173 16 GyroscopeNoise 0.8173 16 AccelerometerBiasNoise 0.8172 16 GyroscopeBiasNoise 0.8172 16 GPSPositionNoise 0.8122 16 GPSVelocityNoise 0.8116 16 MVOPositionNoise 0.7961 16 MVOOrientationNoise 0.7858 17 AccelerometerNoise 0.7858 17 GyroscopeNoise 0.7858 17 AccelerometerBiasNoise 0.7857 17 GyroscopeBiasNoise 0.7857 17 GPSPositionNoise 0.7807 17 GPSVelocityNoise 0.7800 17 MVOPositionNoise 0.7655 17 MVOOrientationNoise 0.7572 18 AccelerometerNoise 0.7572 18 GyroscopeNoise 0.7572 18 AccelerometerBiasNoise 0.7570 18 GyroscopeBiasNoise 0.7570 18 GPSPositionNoise 0.7525 18 GPSVelocityNoise 0.7520 18 MVOPositionNoise 0.7401 18 MVOOrientationNoise 0.7338 19 AccelerometerNoise 0.7337 19 GyroscopeNoise 0.7337 19 AccelerometerBiasNoise 0.7335 19 GyroscopeBiasNoise 0.7335 19 GPSPositionNoise 0.7293 19 GPSVelocityNoise 0.7290 19 MVOPositionNoise 0.7185 19 MVOOrientationNoise 0.7140 20 AccelerometerNoise 0.7138 20 GyroscopeNoise 0.7138 20 AccelerometerBiasNoise 0.7134 20 GyroscopeBiasNoise 0.7134 20 GPSPositionNoise 0.7086 20 GPSVelocityNoise 0.7068 20 MVOPositionNoise 0.6956 20 MVOOrientationNoise 0.6926 21 AccelerometerNoise 0.6922 21 GyroscopeNoise 0.6922 21 AccelerometerBiasNoise 0.6916 21 GyroscopeBiasNoise 0.6916 21 GPSPositionNoise 0.6862 21 GPSVelocityNoise 0.6822 21 MVOPositionNoise 0.6682 21 MVOOrientationNoise 0.6667 22 AccelerometerNoise 0.6660 22 GyroscopeNoise 0.6660 22 AccelerometerBiasNoise 0.6650 22 GyroscopeBiasNoise 0.6650 22 GPSPositionNoise 0.6605 22 GPSVelocityNoise 0.6541 22 MVOPositionNoise 0.6372 22 MVOOrientationNoise 0.6368 23 AccelerometerNoise 0.6356 23 GyroscopeNoise 0.6356 23 AccelerometerBiasNoise 0.6344 23 GyroscopeBiasNoise 0.6344 23 GPSPositionNoise 0.6324 23 GPSVelocityNoise 0.6252 23 MVOPositionNoise 0.6087 23 MVOOrientationNoise 0.6087 24 AccelerometerNoise 0.6075 24 GyroscopeNoise 0.6075 24 AccelerometerBiasNoise 0.6068 24 GyroscopeBiasNoise 0.6068 24 GPSPositionNoise 0.6061 24 GPSVelocityNoise 0.6032 24 MVOPositionNoise 0.6032 24 MVOOrientationNoise 0.6032 25 AccelerometerNoise 0.6017 25 GyroscopeNoise 0.6017 25 AccelerometerBiasNoise 0.6012 25 GyroscopeBiasNoise 0.6012 25 GPSPositionNoise 0.6010 25 GPSVelocityNoise 0.6005 25 MVOPositionNoise 0.6005 25 MVOOrientationNoise 0.6005 26 AccelerometerNoise 0.5992 26 GyroscopeNoise 0.5992 26 AccelerometerBiasNoise 0.5987 26 GyroscopeBiasNoise 0.5987 26 GPSPositionNoise 0.5983 26 GPSVelocityNoise 0.5983 26 MVOPositionNoise 0.5983 26 MVOOrientationNoise 0.5983 27 AccelerometerNoise 0.5975 27 GyroscopeNoise 0.5975 27 AccelerometerBiasNoise 0.5974 27 GyroscopeBiasNoise 0.5974 27 GPSPositionNoise 0.5973 27 GPSVelocityNoise 0.5972 27 MVOPositionNoise 0.5971 27 MVOOrientationNoise 0.5971 28 AccelerometerNoise 0.5971 28 GyroscopeNoise 0.5971 28 AccelerometerBiasNoise 0.5970 28 GyroscopeBiasNoise 0.5970 28 GPSPositionNoise 0.5970 28 GPSVelocityNoise 0.5970 28 MVOPositionNoise 0.5970 28 MVOOrientationNoise 0.5970 29 AccelerometerNoise 0.5970 29 GyroscopeNoise 0.5970 29 AccelerometerBiasNoise 0.5970 29 GyroscopeBiasNoise 0.5970 29 GPSPositionNoise 0.5970 29 GPSVelocityNoise 0.5970 29 MVOPositionNoise 0.5970 29 MVOOrientationNoise 0.5970 30 AccelerometerNoise 0.5969 30 GyroscopeNoise 0.5969 30 AccelerometerBiasNoise 0.5969 30 GyroscopeBiasNoise 0.5969 30 GPSPositionNoise 0.5969 30 GPSVelocityNoise 0.5969 30 MVOPositionNoise 0.5968 30 MVOOrientationNoise 0.5968 31 AccelerometerNoise 0.5968 31 GyroscopeNoise 0.5968 31 AccelerometerBiasNoise 0.5968 31 GyroscopeBiasNoise 0.5968 31 GPSPositionNoise 0.5968 31 GPSVelocityNoise 0.5968 31 MVOPositionNoise 0.5967 31 MVOOrientationNoise 0.5967 32 AccelerometerNoise 0.5967 32 GyroscopeNoise 0.5967 32 AccelerometerBiasNoise 0.5967 32 GyroscopeBiasNoise 0.5967 32 GPSPositionNoise 0.5967 32 GPSVelocityNoise 0.5967 32 MVOPositionNoise 0.5966 32 MVOOrientationNoise 0.5966 33 AccelerometerNoise 0.5966 33 GyroscopeNoise 0.5966 33 AccelerometerBiasNoise 0.5966 33 GyroscopeBiasNoise 0.5966 33 GPSPositionNoise 0.5966 33 GPSVelocityNoise 0.5966 33 MVOPositionNoise 0.5965 33 MVOOrientationNoise 0.5965 34 AccelerometerNoise 0.5965 34 GyroscopeNoise 0.5965 34 AccelerometerBiasNoise 0.5965 34 GyroscopeBiasNoise 0.5965 34 GPSPositionNoise 0.5965 34 GPSVelocityNoise 0.5964 34 MVOPositionNoise 0.5964 34 MVOOrientationNoise 0.5964 35 AccelerometerNoise 0.5964 35 GyroscopeNoise 0.5964 35 AccelerometerBiasNoise 0.5963 35 GyroscopeBiasNoise 0.5963 35 GPSPositionNoise 0.5963 35 GPSVelocityNoise 0.5963 35 MVOPositionNoise 0.5963 35 MVOOrientationNoise 0.5963 36 AccelerometerNoise 0.5963 36 GyroscopeNoise 0.5963 36 AccelerometerBiasNoise 0.5963 36 GyroscopeBiasNoise 0.5963 36 GPSPositionNoise 0.5963 36 GPSVelocityNoise 0.5963 36 MVOPositionNoise 0.5963 36 MVOOrientationNoise 0.5963 37 AccelerometerNoise 0.5963 37 GyroscopeNoise 0.5963 37 AccelerometerBiasNoise 0.5963 37 GyroscopeBiasNoise 0.5963 37 GPSPositionNoise 0.5962 37 GPSVelocityNoise 0.5962 37 MVOPositionNoise 0.5962 37 MVOOrientationNoise 0.5962 38 AccelerometerNoise 0.5962 38 GyroscopeNoise 0.5962 38 AccelerometerBiasNoise 0.5962 38 GyroscopeBiasNoise 0.5962 38 GPSPositionNoise 0.5962 38 GPSVelocityNoise 0.5961 38 MVOPositionNoise 0.5961 38 MVOOrientationNoise 0.5961 39 AccelerometerNoise 0.5961 39 GyroscopeNoise 0.5961 39 AccelerometerBiasNoise 0.5961 39 GyroscopeBiasNoise 0.5961 39 GPSPositionNoise 0.5961 39 GPSVelocityNoise 0.5960 39 MVOPositionNoise 0.5960 39 MVOOrientationNoise 0.5960 40 AccelerometerNoise 0.5960 40 GyroscopeNoise 0.5960 40 AccelerometerBiasNoise 0.5960 40 GyroscopeBiasNoise 0.5960 40 GPSPositionNoise 0.5960 40 GPSVelocityNoise 0.5959 40 MVOPositionNoise 0.5959 40 MVOOrientationNoise 0.5959
Fuse the sensor data using the tuned filter.
N = size(sensorData,1); qEstTuned = quaternion.zeros(N,1); posEstTuned = zeros(N,3); for ii=1:N predict(filter, Accelerometer(ii,:),Gyroscope(ii,:)); if all(~isnan(GPSPosition(ii,1))) fusegps(filter,GPSPosition(ii,:), ... tunedmn.GPSPositionNoise,GPSVelocity(ii,:), ... tunedmn.GPSVelocityNoise); end if all(~isnan(MVOPosition(ii,1))) fusemvo(filter,MVOPosition(ii,:),tunedmn.MVOPositionNoise, ... MVOOrientation{ii},tunedmn.MVOOrientationNoise); end [posEstTuned(ii,:),qEstTuned(ii,:)] = pose(filter); end
Compute the RMS errors.
orientationErrorTuned = rad2deg(dist(qEstTuned,Orientation)); rmsOrientationErrorTuned = sqrt(mean(orientationErrorTuned.^2))
rmsOrientationErrorTuned = 4.4999
positionErrorTuned = sqrt(sum((posEstTuned - Position).^2,2)); rmsPositionErrorTuned = sqrt(mean( positionErrorTuned.^2))
rmsPositionErrorTuned = 0.1172
Visualize the results.
figure; t = (0:N-1)./filter.IMUSampleRate; subplot(2,1,1) plot(t, positionErrorTuned,'b'); title("Tuned insfilterErrorState" + newline + ... "Euclidean Distance Position Error") xlabel('Time (s)'); ylabel('Position Error (meters)') subplot(2,1,2) plot(t, orientationErrorTuned,'b'); title("Orientation Error") xlabel('Time (s)'); ylabel('Orientation Error (degrees)');
Input Arguments
filter
— Filter object
insfilterErrorState
object
Filter object, specified as an insfilterErrorState
object.
measureNoise
— Measurement noise
structure
Measurement noise, specified as a structure. The function uses the measurement noise input as the initial guess for tuning the measurement noise. The structure must contain these fields:
Field name | Description |
---|---|
MVOOrientationNoise | Orientation measurement covariance of monocular visual odometry, specified as a scalar, 3-element vector, or 3-by-3 matrix in rad2 |
MVOPositionNoise | Position measurement covariance of MVO, specified as a scalar, 3-element vector, or 3-by-3 matrix in m2 |
GPSPositionNoise | Variance of GPS position noise, specified as a scalar in m2 |
GPSVelocityNoise | Variance of GPS velocity noise, specified as a scalar in (m/s)2 |
sensorData
— Sensor data
table
Sensor data, specified as a table. In each row, the sensor data is specified as:
Accelerometer
— Accelerometer data, specified as a 1-by-3 vector of scalars in m2/s.Gyroscope
— Gyroscope data, specified as a 1-by-3 vector of scalars in rad/s.MVOOrienation
— Orientation of the camera with respect to the local navigation frame, specified as a scalar quaternion or 3-by-3 rotation matrix. The quaternion or rotation matrix is a frame rotation from the local navigation frame to the current camera coordinate system.MVOPosition
— Position of camera in the local navigation frame, specified as a real 3-element row vector in meters.GPSPosition
— GPS position data, specified as a 1-by-3 vector of latitude in degrees, longitude in degrees, and altitude in meters.GPSVelocity
— GPS velocity data, specified as a 1-by-3 vector of scalars in m/s.
If the GPS sensor does not produce complete measurements, specify the
corresponding entry for GPSPosition
and/or
GPSVelocity
as NaN
. If you set the
Cost
property of the tuner configuration input,
config
, to Custom
, then you can use other data
types for the sensorData
input based on your choice.
groundTruth
— Ground truth data
table
Ground truth data, specified as a table
. In each row, the table
can optionally contain any of these variables:
Orientation
— Orientation from the navigation frame to the body frame, specified as aquaternion
or a 3-by-3 rotation matrix.Position
— Position in navigation frame, specified as a 1-by-3 vector of scalars in meters.Velocity
— Velocity in navigation frame, specified as a 1-by-3 vector of scalars in m/s.AccelerometerBias
— Accelerometer delta angle bias in body frame, specified as a 1-by-3 vector of scalars in m2/s.VisualOdometryScale
— Visual odometry scale factor, specified as a scalar.
The function processes each row of the sensorData
and
groundTruth
tables sequentially to calculate the state estimate
and RMS error from the ground truth. State variables not present in
groundTruth
input are ignored for the comparison. The
sensorData
and the groundTruth
tables must
have the same number of rows.
If you set the Cost
property of the tuner configuration input,
config
, to Custom
, then you can use other data
types for the groundTruth
input based on your choice.
config
— Tuner configuration
tunerconfig
object
Tuner configuration, specified as a tunerconfig
object.
Output Arguments
tunedMeasureNoise
— Tuned measurement noise
structure
Tuned measurement noise, returned as a structure. The structure contains these fields.
Field name | Description |
---|---|
MVOOrientationNoise | Orientation measurement covariance of monocular visual odometry, specified as a scalar, 3-element vector, or 3-by-3 matrix in rad2 |
MVOPositionNoise | Position measurement covariance of MVO, specified as a scalar, 3-element vector, or 3-by-3 matrix in m2 |
GPSPositionNoise | Variance of GPS position noise, specified as a scalar in m2 |
GPSVelocityNoise | Variance of GPS velocity noise, specified as a scalar in (m/s)2 |
References
[1] Abbeel, P., Coates, A., Montemerlo, M., Ng, A.Y. and Thrun, S. Discriminative Training of Kalman Filters. In Robotics: Science and systems, Vol. 2, pp. 1, 2005.
Version History
Introduced in R2021a
See Also
Commande MATLAB
Vous avez cliqué sur un lien qui correspond à cette commande MATLAB :
Pour exécuter la commande, saisissez-la dans la fenêtre de commande de MATLAB. Les navigateurs web ne supportent pas les commandes MATLAB.
Select a Web Site
Choose a web site to get translated content where available and see local events and offers. Based on your location, we recommend that you select: .
You can also select a web site from the following list:
How to Get Best Site Performance
Select the China site (in Chinese or English) for best site performance. Other MathWorks country sites are not optimized for visits from your location.
Americas
- América Latina (Español)
- Canada (English)
- United States (English)
Europe
- Belgium (English)
- Denmark (English)
- Deutschland (Deutsch)
- España (Español)
- Finland (English)
- France (Français)
- Ireland (English)
- Italia (Italiano)
- Luxembourg (English)
- Netherlands (English)
- Norway (English)
- Österreich (Deutsch)
- Portugal (English)
- Sweden (English)
- Switzerland
- United Kingdom (English)