Package com.irurueta.navigation.inertial.calibration.accelerometer

Class RobustKnownPositionAccelerometerCalibrator.PreliminaryResult

java.lang.Object

com.irurueta.navigation.inertial.calibration.accelerometer.RobustKnownPositionAccelerometerCalibrator.PreliminaryResult

Enclosing class:

RobustKnownPositionAccelerometerCalibrator

protected static class RobustKnownPositionAccelerometerCalibrator.PreliminaryResult
extends Object

Internal class containing estimated preliminary result.

Field Summary

Fields

Modifier and Type	Field	Description
private com.irurueta.algebra.Matrix	covariance	Estimated covariance matrix.
private double[]	estimatedBiases	Estimated accelerometer biases for each IMU axis expressed in meter per squared second (m/s^2).
private double	estimatedChiSq	Estimated chi square value.
private com.irurueta.algebra.Matrix	estimatedMa	Estimated accelerometer scale factors and cross coupling errors.
private double	estimatedMse	Estimated MSE (Mean Square Error).

Constructor Summary

Constructors

Modifier	Constructor	Description
protected	PreliminaryResult()

Method Summary

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Field Details

estimatedBiases

private double[] estimatedBiases

Estimated accelerometer biases for each IMU axis expressed in meter per squared second (m/s^2).

estimatedMa

private com.irurueta.algebra.Matrix estimatedMa

Estimated accelerometer scale factors and cross coupling errors. This is the product of matrix Ta containing cross coupling errors and Ka containing scaling factors. So tat:

     Ma = [sx    mxy  mxz] = Ta*Ka
          [myx   sy   myz]
          [mzx   mzy  sz ]
 

Where:

     Ka = [sx 0   0 ]
          [0  sy  0 ]
          [0  0   sz]
 

and

     Ta = [1          -alphaXy    alphaXz ]
          [alphaYx    1           -alphaYz]
          [-alphaZx   alphaZy     1       ]
 

Hence:

     Ma = [sx    mxy  mxz] = Ta*Ka =  [sx             -sy * alphaXy   sz * alphaXz ]
          [myx   sy   myz]            [sx * alphaYx   sy              -sz * alphaYz]
          [mzx   mzy  sz ]            [-sx * alphaZx  sy * alphaZy    sz           ]
 

This instance allows any 3x3 matrix however, typically alphaYx, alphaZx and alphaZy are considered to be zero if the accelerometer z-axis is assumed to be the same as the body z-axis. When this is assumed, myx = mzx = mzy = 0 and the Ma matrix becomes upper diagonal:

     Ma = [sx    mxy  mxz]
          [0     sy   myz]
          [0     0    sz ]
 

Values of this matrix are unit-less.

covariance

private com.irurueta.algebra.Matrix covariance

Estimated covariance matrix.

estimatedMse

private double estimatedMse

Estimated MSE (Mean Square Error).

estimatedChiSq

private double estimatedChiSq

Estimated chi square value.

Constructor Details

PreliminaryResult

protected PreliminaryResult()