Lie groups
Host-side rigid-body helpers (Mink-compatible). Device kernels live in
mink_warp.lie.wp_ops.
-
class mink_warp.SE3[source]
Bases: object
Special Euclidean group for proper rigid transforms in 3D.
Internal parameterization is (qw, qx, qy, qz, x, y, z). Tangent parameterization is
(vx, vy, vz, omega_x, omega_y, omega_z). Matches Mink’s SE3.
-
wxyz_xyz: ndarray
-
matrix_dim: int = 4
-
parameters_dim: int = 7
-
tangent_dim: int = 6
-
space_dim: int = 3
-
copy() → SE3[source]
-
parameters() → ndarray[source]
-
classmethod identity() → SE3[source]
-
classmethod from_rotation_and_translation(rotation: SO3, translation: ndarray) → SE3[source]
-
classmethod from_rotation(rotation: SO3) → SE3[source]
-
classmethod from_translation(translation: ndarray) → SE3[source]
-
classmethod from_matrix(matrix: ndarray) → SE3[source]
-
classmethod sample_uniform() → SE3[source]
-
rotation() → SO3[source]
-
translation() → ndarray[source]
-
as_matrix() → ndarray[source]
-
classmethod exp(tangent: ndarray) → SE3[source]
-
inverse() → SE3[source]
-
normalize() → SE3[source]
-
apply(target: ndarray) → ndarray[source]
-
multiply(other: SE3) → SE3[source]
-
log() → ndarray[source]
-
adjoint() → ndarray[source]
-
rplus(other: ndarray) → SE3[source]
-
rminus(other: SE3) → ndarray[source]
-
plus(other: ndarray) → SE3[source]
-
minus(other: SE3) → ndarray[source]
-
classmethod ljac(other: ndarray) → ndarray[source]
-
classmethod ljacinv(other: ndarray) → ndarray[source]
-
classmethod rjac(other: ndarray) → ndarray[source]
-
classmethod rjacinv(other: ndarray) → ndarray[source]
-
jlog() → ndarray[source]
-
class mink_warp.SO3[source]
Bases: object
Special orthogonal group for 3D rotations.
Internal parameterization is (qw, qx, qy, qz). Tangent parameterization is
(omega_x, omega_y, omega_z). Matches Mink’s SO3.
-
wxyz: ndarray
-
matrix_dim: int = 3
-
parameters_dim: int = 4
-
tangent_dim: int = 3
-
space_dim: int = 3
-
parameters() → ndarray[source]
-
copy() → SO3[source]
-
classmethod from_matrix(matrix: ndarray) → SO3[source]
-
classmethod identity() → SO3[source]
-
as_matrix() → ndarray[source]
-
inverse() → SO3[source]
-
normalize() → SO3[source]
-
apply(target: ndarray) → ndarray[source]
-
multiply(other: SO3) → SO3[source]
-
classmethod exp(tangent: ndarray) → SO3[source]
-
log() → ndarray[source]
-
adjoint() → ndarray[source]
-
classmethod ljac(other: ndarray) → ndarray[source]
-
classmethod ljacinv(other: ndarray) → ndarray[source]
-
classmethod rjac(other: ndarray) → ndarray[source]
-
classmethod rjacinv(other: ndarray) → ndarray[source]
-
jlog() → ndarray[source]