Meaning
Mathematical arrays of joint variance define the multi-dimensional dispersion of coordinate uncertainties in a spatial measurement system. In spatial metrology, a covariance matrix represents the statistical dependencies and individual variances of measured point coordinates. Each diagonal element of the array specifies the variance of a single coordinate axis, whereas off-diagonal elements quantify the correlation between different axes.
Calculations of coordinate uncertainty rely on these arrays to propagate errors when transforming points between coordinate frames.
Error Propagation
Linear transformations of spatial coordinates map uncertainty from one system to another through multiplication with Jacobian matrices. In this process, the covariance matrix undergoes a transformation that projects the error ellipsoid into the new reference frame. A coordinate rotation redistributes the variance across the new axes, which alters the correlation coefficients.
Accurate projection preserves the geometric relationships of the measurement uncertainty.
Coordinate Transformation
Transforming measurements across different coordinate systems demands a consistent mathematical structure to avoid skewing error estimates. The transformed covariance matrix ensures that the uncertainty ellipsoid retains its physical size and orientation regardless of the selected origin or axis direction. Laser trackers utilize this mechanism to maintain spatial consistency during alignment.
When multiple instruments coordinate their measurements, this transformation prevents the distortion of the combined statistical models.
Precision Estimate
An estimation of positional precision relies on extracting the eigenvalues of the array to define the axes of the uncertainty ellipsoid. This precision estimate reveals the directions of highest and lowest measurement confidence. High confidence corresponds to the shortest axis of the ellipsoid.