Edge constraints in pose graph
measurements— Measurements between nodes
Measurements between nodes, returned as an n-by-3 matrix or n-by-7 matrix.
poseGraph (2-D), each row is an
[x y theta]
vector, which defines the relative xy-position and
theta, of a pose in the graph. For
theta is returned as
poseGraph3D, each row is an
[x y z qw qx qy
qz] vector, which defines the relative
xyz-position and quaternion orientation,
qy qz], of a pose in the graph.
Many other sources for 3-D pose graphs, including
.g2o formats, specify the quaternion orientation
in a different order, for example,
[qx qy qz qw].
Check the source of your pose graph data before adding nodes to your
infoMats— Information matrices
Information matrices, specified in compact form as a n-by-6 or n-by-21 matrix, where n is the number of poses in the pose graph.
Each row is the upper triangle of the square information matrix. An
information matrix represents the uncertainty of the measurement. The matrix
is calculated as the inverse of the covariance. If the measurement is an
[x y theta] vector, the covariance matrix is a 3-by-3
of pairwise covariance calculations. Typically, the uncertainty is
determined by the sensor model.
poseGraph (2-D), each information matrix is a six-element
vector. The default is
[1 0 0 1 0 1]. For landmark nodes,
the last three elements are returned as
poseGraph3D, each information matrix is a 21-element vector.
The default is
[1 0 0 0 0 0 1 0 0 0 0 1 0 0 0 1 0 0 1 0
Use this syntax when constructing
poseGraph3D objects for code generation:
poseGraph('MaxNumEdges',maxEdges,'MaxNumNodes',maxNodes) specifies an
upper bound on the number of edges and nodes allowed in the pose graph when
generating code. This limit is only required when generating code.