pytransform3d.uncertainty.to_projected_ellipsoid#

pytransform3d.uncertainty.to_projected_ellipsoid(mean, cov, factor=1.96, n_steps=20)[source]#

Compute projected error ellipsoid.

An error ellipsoid shows equiprobable points. This is a projection of a Gaussian distribution in exponential coordinate space to 3D.

Parameters:
meanarray-like, shape (4, 4)

Mean of pose distribution.

covarray-like, shape (6, 6)

Covariance of pose distribution in exponential coordinate space.

factorfloat, optional (default: 1.96)

Multiple of the standard deviations that should be plotted.

n_stepsint, optional (default: 20)

Number of discrete steps plotted in each dimension.

Returns:
xarray, shape (n_steps, n_steps)

Coordinates on x-axis of grid on projected ellipsoid.

yarray, shape (n_steps, n_steps)

Coordinates on y-axis of grid on projected ellipsoid.

zarray, shape (n_steps, n_steps)

Coordinates on z-axis of grid on projected ellipsoid.

Examples using pytransform3d.uncertainty.to_projected_ellipsoid#

Probabilistic Product of Exponentials

Probabilistic Product of Exponentials