Matrix Condition
Failure of a calibration model to distinguish between independent error sources due to insufficient or redundant data. Scale factor matrix degeneracy happens when the test positions do not provide enough unique information to solve for every variable in the sensor equation. This results in a matrix that cannot be inverted or produces unstable solutions.
The system becomes unable to separate the scale factor from the misalignment.
Test Coverage
Geometry of the sampling points dictates the health of the calibration routine. To prevent scale factor matrix degeneracy, the rotation sequence must include movements that stimulate each axis of the sensor in multiple directions.
Numerical Stability
Performance of the matrix inversion depends on the condition number of the data set. High values indicate that the scale factor matrix degeneracy is imminent, meaning small measurement noises will cause large swings in the calculated coefficients. Monitoring this value during calibration ensures that the resulting model is stable and repeatable.
Compensation Error
Application of a degenerate model leads to measurable inaccuracies in the field. When scale factor matrix degeneracy is present, the device may report correct values in the test positions but fail at intermediate angles. This error appears as a periodic distortion that degrades the overall navigation or sensing solution.
