LMedSRobustKnownFrameGyroscopeCalibrator.java
/*
* Copyright (C) 2020 Alberto Irurueta Carro (alberto@irurueta.com)
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package com.irurueta.navigation.inertial.calibration.gyroscope;
import com.irurueta.navigation.LockedException;
import com.irurueta.navigation.NotReadyException;
import com.irurueta.navigation.inertial.calibration.CalibrationException;
import com.irurueta.navigation.inertial.calibration.StandardDeviationFrameBodyKinematics;
import com.irurueta.numerical.robust.LMedSRobustEstimator;
import com.irurueta.numerical.robust.LMedSRobustEstimatorListener;
import com.irurueta.numerical.robust.RobustEstimator;
import com.irurueta.numerical.robust.RobustEstimatorException;
import com.irurueta.numerical.robust.RobustEstimatorMethod;
import java.util.List;
/**
* Robustly estimates gyroscope biases, cross couplings and scaling factors
* along with G-dependent cross biases introduced on the gyroscope by the
* specific forces sensed by the accelerometer using an LMedS algorithm to discard
* outliers.
* <p>
* To use this calibrator at least 7 measurements at different known frames must
* be provided. In other words, accelerometer and gyroscope (i.e. body kinematics)
* samples must be obtained at 7 different positions, orientations and velocities
* (although typically velocities are always zero).
* <p>
* Measured gyroscope angular rates is assumed to follow the model shown below:
* <pre>
* Ωmeas = bg + (I + Mg) * Ωtrue + Gg * ftrue + w
* </pre>
* Where:
* - Ωmeas is the measured gyroscope angular rates. This is a 3x1 vector.
* - bg is the gyroscope bias. Ideally, on a perfect gyroscope, this should be a
* 3x1 zero vector.
* - I is the 3x3 identity matrix.
* - Mg is the 3x3 matrix containing cross-couplings and scaling factors. Ideally, on
* a perfect gyroscope, this should be a 3x3 zero matrix.
* - Ωtrue is ground-truth gyroscope angular rates.
* - Gg is the G-dependent cross biases introduced by the specific forces sensed
* by the accelerometer. Ideally, on a perfect gyroscope, this should be a 3x3
* zero matrix.
* - ftrue is ground-truth specific force. This is a 3x1 vector.
* - w is measurement noise. This is a 3x1 vector.
*/
public class LMedSRobustKnownFrameGyroscopeCalibrator extends RobustKnownFrameGyroscopeCalibrator {
/**
* Default value to be used for stop threshold. Stop threshold can be used to
* avoid keeping the algorithm unnecessarily iterating in case that best
* estimated threshold using median of residuals is not small enough. Once a
* solution is found that generates a threshold below this value, the
* algorithm will stop.
* The stop threshold can be used to prevent the LMedS algorithm iterating
* too many times in cases where samples have a very similar accuracy.
* For instance, in cases where proportion of outliers is very small (close
* to 0%), and samples are very accurate (i.e. 1e-6), the algorithm would
* iterate for a long time trying to find the best solution when indeed
* there is no need to do that if a reasonable threshold has already been
* reached.
* Because of this behaviour the stop threshold can be set to a value much
* lower than the one typically used in RANSAC, and yet the algorithm could
* still produce even smaller thresholds in estimated results.
*/
public static final double DEFAULT_STOP_THRESHOLD = 5e-4;
/**
* Minimum allowed stop threshold value.
*/
public static final double MIN_STOP_THRESHOLD = 0.0;
/**
* Threshold to be used to keep the algorithm iterating in case that best
* estimated threshold using median of residuals is not small enough. Once
* a solution is found that generates a threshold below this value, the
* algorithm will stop.
* The stop threshold can be used to prevent the LMedS algorithm iterating
* too many times in cases where samples have a very similar accuracy.
* For instance, in cases where proportion of outliers is very small (close
* to 0%), and samples are very accurate (i.e. 1e-6), the algorithm would
* iterate for a long time trying to find the best solution when indeed
* there is no need to do that if a reasonable threshold has already been
* reached.
* Because of this behaviour the stop threshold can be set to a value much
* lower than the one typically used in RANSAC, and yet the algorithm could
* still produce even smaller thresholds in estimated results.
*/
private double stopThreshold = DEFAULT_STOP_THRESHOLD;
/**
* Constructor.
*/
public LMedSRobustKnownFrameGyroscopeCalibrator() {
}
/**
* Constructor.
*
* @param listener listener to be notified of events such as when estimation
* starts, ends or its progress significantly changes.
*/
public LMedSRobustKnownFrameGyroscopeCalibrator(final RobustKnownFrameGyroscopeCalibratorListener listener) {
super(listener);
}
/**
* Constructor.
*
* @param measurements list of body kinematics measurements with standard
* deviations taken at different frames (positions, orientations
* and velocities).
*/
public LMedSRobustKnownFrameGyroscopeCalibrator(final List<StandardDeviationFrameBodyKinematics> measurements) {
super(measurements);
}
/**
* Constructor.
*
* @param measurements list of body kinematics measurements with standard
* deviations taken at different frames (positions, orientations
* and velocities).
* @param listener listener to be notified of events such as when estimation
* starts, ends or its progress significantly changes.
*/
public LMedSRobustKnownFrameGyroscopeCalibrator(
final List<StandardDeviationFrameBodyKinematics> measurements,
final RobustKnownFrameGyroscopeCalibratorListener listener) {
super(measurements, listener);
}
/**
* Constructor.
*
* @param commonAxisUsed indicates whether z-axis is assumed to be common for
* accelerometer and gyroscope.
*/
public LMedSRobustKnownFrameGyroscopeCalibrator(final boolean commonAxisUsed) {
super(commonAxisUsed);
}
/**
* Constructor.
*
* @param commonAxisUsed indicates whether z-axis is assumed to be common for
* accelerometer and gyroscope.
* @param listener listener to handle events raised by this calibrator.
*/
public LMedSRobustKnownFrameGyroscopeCalibrator(
final boolean commonAxisUsed, final RobustKnownFrameGyroscopeCalibratorListener listener) {
super(commonAxisUsed, listener);
}
/**
* Constructor.
*
* @param measurements list of body kinematics measurements with standard
* deviations taken at different frames (positions, orientations
* and velocities).
* @param commonAxisUsed indicates whether z-axis is assumed to be common for
* accelerometer and gyroscope.
*/
public LMedSRobustKnownFrameGyroscopeCalibrator(
final List<StandardDeviationFrameBodyKinematics> measurements, final boolean commonAxisUsed) {
super(measurements, commonAxisUsed);
}
/**
* Constructor.
*
* @param measurements list of body kinematics measurements with standard
* deviations taken at different frames (positions, orientations
* and velocities).
* @param commonAxisUsed indicates whether z-axis is assumed to be common for
* accelerometer and gyroscope.
* @param listener listener to handle events raised by this calibrator.
*/
public LMedSRobustKnownFrameGyroscopeCalibrator(
final List<StandardDeviationFrameBodyKinematics> measurements, final boolean commonAxisUsed,
final RobustKnownFrameGyroscopeCalibratorListener listener) {
super(measurements, commonAxisUsed, listener);
}
/**
* Returns threshold to be used to keep the algorithm iterating in case that
* best estimated threshold using median of residuals is not small enough.
* Once a solution is found that generates a threshold below this value, the
* algorithm will stop.
* The stop threshold can be used to prevent the LMedS algorithm to iterate
* too many times in cases where samples have a very similar accuracy.
* For instance, in cases where proportion of outliers is very small (close
* to 0%), and samples are very accurate (i.e. 1e-6), the algorithm would
* iterate for a long time trying to find the best solution when indeed
* there is no need to do that if a reasonable threshold has already been
* reached.
* Because of this behaviour the stop threshold can be set to a value much
* lower than the one typically used in RANSAC, and yet the algorithm could
* still produce even smaller thresholds in estimated results.
*
* @return stop threshold to stop the algorithm prematurely when a certain
* accuracy has been reached.
*/
public double getStopThreshold() {
return stopThreshold;
}
/**
* Sets threshold to be used to keep the algorithm iterating in case that
* best estimated threshold using median of residuals is not small enough.
* Once a solution is found that generates a threshold below this value,
* the algorithm will stop.
* The stop threshold can be used to prevent the LMedS algorithm to iterate
* too many times in cases where samples have a very similar accuracy.
* For instance, in cases where proportion of outliers is very small (close
* to 0%), and samples are very accurate (i.e. 1e-6), the algorithm would
* iterate for a long time trying to find the best solution when indeed
* there is no need to do that if a reasonable threshold has already been
* reached.
* Because of this behaviour the stop threshold can be set to a value much
* lower than the one typically used in RANSAC, and yet the algorithm could
* still produce even smaller thresholds in estimated results.
*
* @param stopThreshold stop threshold to stop the algorithm prematurely
* when a certain accuracy has been reached.
* @throws IllegalArgumentException if provided value is zero or negative.
* @throws LockedException if calibrator is currently running.
*/
public void setStopThreshold(final double stopThreshold) throws LockedException {
if (running) {
throw new LockedException();
}
if (stopThreshold <= MIN_STOP_THRESHOLD) {
throw new IllegalArgumentException();
}
this.stopThreshold = stopThreshold;
}
/**
* Estimates accelerometer calibration parameters containing bias, scale factors
* and cross-coupling errors.
*
* @throws LockedException if calibrator is currently running.
* @throws NotReadyException if calibrator is not ready.
* @throws CalibrationException if estimation fails for numerical reasons.
*/
@SuppressWarnings("DuplicatedCode")
@Override
public void calibrate() throws LockedException, NotReadyException, CalibrationException {
if (running) {
throw new LockedException();
}
if (!isReady()) {
throw new NotReadyException();
}
final var innerEstimator = new LMedSRobustEstimator<>(new LMedSRobustEstimatorListener<PreliminaryResult>() {
@Override
public int getTotalSamples() {
return measurements.size();
}
@Override
public int getSubsetSize() {
return preliminarySubsetSize;
}
@Override
public void estimatePreliminarSolutions(
final int[] samplesIndices, final List<PreliminaryResult> solutions) {
computePreliminarySolutions(samplesIndices, solutions);
}
@Override
public double computeResidual(final PreliminaryResult currentEstimation, final int i) {
return computeError(measurements.get(i), currentEstimation);
}
@Override
public boolean isReady() {
return LMedSRobustKnownFrameGyroscopeCalibrator.super.isReady();
}
@Override
public void onEstimateStart(final RobustEstimator<PreliminaryResult> estimator) {
// no action needed
}
@Override
public void onEstimateEnd(final RobustEstimator<PreliminaryResult> estimator) {
// no action needed
}
@Override
public void onEstimateNextIteration(
final RobustEstimator<PreliminaryResult> estimator, final int iteration) {
if (listener != null) {
listener.onCalibrateNextIteration(
LMedSRobustKnownFrameGyroscopeCalibrator.this, iteration);
}
}
@Override
public void onEstimateProgressChange(
final RobustEstimator<PreliminaryResult> estimator, final float progress) {
if (listener != null) {
listener.onCalibrateProgressChange(
LMedSRobustKnownFrameGyroscopeCalibrator.this, progress);
}
}
});
try {
running = true;
if (listener != null) {
listener.onCalibrateStart(this);
}
inliersData = null;
innerEstimator.setConfidence(confidence);
innerEstimator.setMaxIterations(maxIterations);
innerEstimator.setProgressDelta(progressDelta);
innerEstimator.setStopThreshold(stopThreshold);
final var preliminaryResult = innerEstimator.estimate();
inliersData = innerEstimator.getInliersData();
attemptRefine(preliminaryResult);
if (listener != null) {
listener.onCalibrateEnd(this);
}
} catch (final com.irurueta.numerical.LockedException e) {
throw new LockedException(e);
} catch (final com.irurueta.numerical.NotReadyException e) {
throw new NotReadyException(e);
} catch (final RobustEstimatorException e) {
throw new CalibrationException(e);
} finally {
running = false;
}
}
/**
* Returns method being used for robust estimation.
*
* @return method being used for robust estimation.
*/
@Override
public RobustEstimatorMethod getMethod() {
return RobustEstimatorMethod.LMEDS;
}
/**
* Indicates whether this calibrator requires quality scores for each
* measurement/sequence or not.
*
* @return true if quality scores are required, false otherwise.
*/
@Override
public boolean isQualityScoresRequired() {
return false;
}
}