`mrpt.graphs`

mrpt.graphs — Pose graph types and graph algorithms.

Provides:

CNetworkOfPoses2D : 2D pose graph (CPose2D edges)
CNetworkOfPoses3D : 3D pose graph (CPose3D edges)
CNetworkOfPoses2DCov, CNetworkOfPoses3DCov : pose graphs with Gaussian edges
CNetworkOfPoses2DInf, CNetworkOfPoses3DInf : pose graphs with edges in information form

Example:

import mrpt.graphs as graphs
import mrpt.poses as poses

g = graphs.CNetworkOfPoses2D()
g.setNodePose(0, poses.CPose2D(0, 0, 0))
g.setNodePose(1, poses.CPose2D(1, 0, 0))
g.insertEdge(0, 1, poses.CPose2D(1, 0, 0))
print(g)

Package Contents

class mrpt.graphs.CNetworkOfPoses2D

A graph of 2D poses (nodes) and relative pose constraints (edges).

property root: int: Root node ID (default: 0)

clearEdges() → None: Erases all the edges.

dijkstra_nodes_estimate() → None: Recomputes all node poses by composing edges along the shortest-path spanning tree from the root node

dijkstra_path(source: int, target: int) → list[int]: Shortest path (fewest edges) between two nodes as a list of node IDs, from source to target, ignoring edge directions. Raises ValueError if target is unreachable.

edgeCount() → int: Number of edges in the graph

edgeExists(from_id: int, to_id: int) → bool: Returns true if the directed edge from_id -> to_id exists.

eraseEdge(from_id: int, to_id: int) → None: Erases all the edges from_id -> to_id (nothing happens if none exists).

getEdge(from_id: int, to_id: int) → mrpt.poses.CPose2D: Returns a copy of the edge from_id -> to_id (the first one, if there are several). Raises KeyError if it does not exist.

getEdges() → list[tuple]: Returns all the edges as a list of tuples (from_id, to_id, edge), sorted by node IDs.

getNeighborsOf(node_id: int) → set[int]: IDs of the nodes connected to the given one by an edge (in any direction)

getNodeDistances(source: int) → dict[int, float]: Topological distance (number of edges) from source to every reachable node

getNodeIDs() → list[int]: Return a list of all node IDs

getNodePose(node_id: int) → mrpt.poses.CPose2D: Get the estimated pose for a node

hasNode(node_id: int) → bool: True if a node with the given ID exists

insertEdge(from_id: int, to_id: int, edge: mrpt.poses.CPose2D) → None: Insert a directed edge from → to

loadFromTextFile(fileName: str, collapse_dup_edges: bool = True) → None: Loads the graph from a text file in the TORO / g2o format.

nodeCount() → int: Number of nodes in the graph

saveToTextFile(fileName: str) → None: Saves the graph to a text file in the TORO / g2o format.

setNodePose(node_id: int, pose: mrpt.poses.CPose2D) → None: Set the estimated pose for a node

class mrpt.graphs.CNetworkOfPoses2DCov

A graph of 2D poses (nodes, CPose2D) and relative pose constraints with a covariance matrix (edges, CPosePDFGaussian).

property root: int: Root node ID (default: 0)

clearEdges() → None: Erases all the edges.

dijkstra_nodes_estimate() → None: Recomputes all node poses by composing edges along the shortest-path spanning tree from the root node

dijkstra_path(source: int, target: int) → list[int]: Shortest path (fewest edges) between two nodes as a list of node IDs, from source to target, ignoring edge directions. Raises ValueError if target is unreachable.

edgeCount() → int: Number of edges in the graph

edgeExists(from_id: int, to_id: int) → bool: Returns true if the directed edge from_id -> to_id exists.

eraseEdge(from_id: int, to_id: int) → None: Erases all the edges from_id -> to_id (nothing happens if none exists).

getEdge(from_id: int, to_id: int) → mrpt.poses.CPosePDFGaussian: Returns a copy of the edge from_id -> to_id (the first one, if there are several). Raises KeyError if it does not exist.

getEdges() → list[tuple]: Returns all the edges as a list of tuples (from_id, to_id, edge), sorted by node IDs.

getNeighborsOf(node_id: int) → set[int]: IDs of the nodes connected to the given one by an edge (in any direction)

getNodeDistances(source: int) → dict[int, float]: Topological distance (number of edges) from source to every reachable node

getNodeIDs() → list[int]: Return a list of all node IDs

getNodePose(node_id: int) → mrpt.poses.CPose2D: Get the estimated pose for a node

hasNode(node_id: int) → bool: True if a node with the given ID exists

insertEdge(from_id: int, to_id: int, edge: mrpt.poses.CPosePDFGaussian) → None: Insert a directed edge from → to

loadFromTextFile(fileName: str, collapse_dup_edges: bool = True) → None: Loads the graph from a text file in the TORO / g2o format.

nodeCount() → int: Number of nodes in the graph

saveToTextFile(fileName: str) → None: Saves the graph to a text file in the TORO / g2o format.

setNodePose(node_id: int, pose: mrpt.poses.CPose2D) → None: Set the estimated pose for a node

class mrpt.graphs.CNetworkOfPoses2DInf

A graph of 2D poses (nodes, CPose2D) and relative pose constraints with an information matrix (edges, CPosePDFGaussianInf).

property root: int: Root node ID (default: 0)

clearEdges() → None: Erases all the edges.

dijkstra_nodes_estimate() → None: Recomputes all node poses by composing edges along the shortest-path spanning tree from the root node

dijkstra_path(source: int, target: int) → list[int]: Shortest path (fewest edges) between two nodes as a list of node IDs, from source to target, ignoring edge directions. Raises ValueError if target is unreachable.

edgeCount() → int: Number of edges in the graph

edgeExists(from_id: int, to_id: int) → bool: Returns true if the directed edge from_id -> to_id exists.

eraseEdge(from_id: int, to_id: int) → None: Erases all the edges from_id -> to_id (nothing happens if none exists).

getEdge(from_id: int, to_id: int) → mrpt.poses.CPosePDFGaussianInf: Returns a copy of the edge from_id -> to_id (the first one, if there are several). Raises KeyError if it does not exist.

getEdges() → list[tuple]: Returns all the edges as a list of tuples (from_id, to_id, edge), sorted by node IDs.

getNeighborsOf(node_id: int) → set[int]: IDs of the nodes connected to the given one by an edge (in any direction)

getNodeDistances(source: int) → dict[int, float]: Topological distance (number of edges) from source to every reachable node

getNodeIDs() → list[int]: Return a list of all node IDs

getNodePose(node_id: int) → mrpt.poses.CPose2D: Get the estimated pose for a node

hasNode(node_id: int) → bool: True if a node with the given ID exists

insertEdge(from_id: int, to_id: int, edge: mrpt.poses.CPosePDFGaussianInf) → None: Insert a directed edge from → to

loadFromTextFile(fileName: str, collapse_dup_edges: bool = True) → None: Loads the graph from a text file in the TORO / g2o format.

nodeCount() → int: Number of nodes in the graph

saveToTextFile(fileName: str) → None: Saves the graph to a text file in the TORO / g2o format.

setNodePose(node_id: int, pose: mrpt.poses.CPose2D) → None: Set the estimated pose for a node

class mrpt.graphs.CNetworkOfPoses3D

A graph of 3D poses (nodes) and relative pose constraints (edges).

property root: int: Root node ID (default: 0)

clearEdges() → None: Erases all the edges.

dijkstra_nodes_estimate() → None: Recomputes all node poses by composing edges along the shortest-path spanning tree from the root node

dijkstra_path(source: int, target: int) → list[int]: Shortest path (fewest edges) between two nodes as a list of node IDs, from source to target, ignoring edge directions. Raises ValueError if target is unreachable.

edgeCount() → int: Number of edges in the graph

edgeExists(from_id: int, to_id: int) → bool: Returns true if the directed edge from_id -> to_id exists.

eraseEdge(from_id: int, to_id: int) → None: Erases all the edges from_id -> to_id (nothing happens if none exists).

getEdge(from_id: int, to_id: int) → mrpt.poses.CPose3D: Returns a copy of the edge from_id -> to_id (the first one, if there are several). Raises KeyError if it does not exist.

getEdges() → list[tuple]: Returns all the edges as a list of tuples (from_id, to_id, edge), sorted by node IDs.

getNeighborsOf(node_id: int) → set[int]: IDs of the nodes connected to the given one by an edge (in any direction)

getNodeDistances(source: int) → dict[int, float]: Topological distance (number of edges) from source to every reachable node

getNodeIDs() → list[int]: Return a list of all node IDs

getNodePose(node_id: int) → mrpt.poses.CPose3D: Get the estimated pose for a node

hasNode(node_id: int) → bool: True if a node with the given ID exists

insertEdge(from_id: int, to_id: int, edge: mrpt.poses.CPose3D) → None: Insert a directed edge from → to

loadFromTextFile(fileName: str, collapse_dup_edges: bool = True) → None: Loads the graph from a text file in the TORO / g2o format.

nodeCount() → int: Number of nodes in the graph

saveToTextFile(fileName: str) → None: Saves the graph to a text file in the TORO / g2o format.

setNodePose(node_id: int, pose: mrpt.poses.CPose3D) → None: Set the estimated pose for a node

class mrpt.graphs.CNetworkOfPoses3DCov

A graph of 3D poses (nodes, CPose3D) and relative pose constraints with a covariance matrix (edges, CPose3DPDFGaussian).

property root: int: Root node ID (default: 0)

clearEdges() → None: Erases all the edges.

dijkstra_nodes_estimate() → None: Recomputes all node poses by composing edges along the shortest-path spanning tree from the root node

dijkstra_path(source: int, target: int) → list[int]: Shortest path (fewest edges) between two nodes as a list of node IDs, from source to target, ignoring edge directions. Raises ValueError if target is unreachable.

edgeCount() → int: Number of edges in the graph

edgeExists(from_id: int, to_id: int) → bool: Returns true if the directed edge from_id -> to_id exists.

eraseEdge(from_id: int, to_id: int) → None: Erases all the edges from_id -> to_id (nothing happens if none exists).

getEdge(from_id: int, to_id: int) → mrpt.poses.CPose3DPDFGaussian: Returns a copy of the edge from_id -> to_id (the first one, if there are several). Raises KeyError if it does not exist.

getEdges() → list[tuple]: Returns all the edges as a list of tuples (from_id, to_id, edge), sorted by node IDs.

getNeighborsOf(node_id: int) → set[int]: IDs of the nodes connected to the given one by an edge (in any direction)

getNodeDistances(source: int) → dict[int, float]: Topological distance (number of edges) from source to every reachable node

getNodeIDs() → list[int]: Return a list of all node IDs

getNodePose(node_id: int) → mrpt.poses.CPose3D: Get the estimated pose for a node

hasNode(node_id: int) → bool: True if a node with the given ID exists

insertEdge(from_id: int, to_id: int, edge: mrpt.poses.CPose3DPDFGaussian) → None: Insert a directed edge from → to

loadFromTextFile(fileName: str, collapse_dup_edges: bool = True) → None: Loads the graph from a text file in the TORO / g2o format.

nodeCount() → int: Number of nodes in the graph

saveToTextFile(fileName: str) → None: Saves the graph to a text file in the TORO / g2o format.

setNodePose(node_id: int, pose: mrpt.poses.CPose3D) → None: Set the estimated pose for a node

class mrpt.graphs.CNetworkOfPoses3DInf

A graph of 3D poses (nodes, CPose3D) and relative pose constraints with an information matrix (edges, CPose3DPDFGaussianInf).

property root: int: Root node ID (default: 0)

clearEdges() → None: Erases all the edges.

dijkstra_nodes_estimate() → None: Recomputes all node poses by composing edges along the shortest-path spanning tree from the root node

dijkstra_path(source: int, target: int) → list[int]: Shortest path (fewest edges) between two nodes as a list of node IDs, from source to target, ignoring edge directions. Raises ValueError if target is unreachable.

edgeCount() → int: Number of edges in the graph

edgeExists(from_id: int, to_id: int) → bool: Returns true if the directed edge from_id -> to_id exists.

eraseEdge(from_id: int, to_id: int) → None: Erases all the edges from_id -> to_id (nothing happens if none exists).

getEdge(from_id: int, to_id: int) → mrpt.poses.CPose3DPDFGaussianInf: Returns a copy of the edge from_id -> to_id (the first one, if there are several). Raises KeyError if it does not exist.

getEdges() → list[tuple]: Returns all the edges as a list of tuples (from_id, to_id, edge), sorted by node IDs.

getNeighborsOf(node_id: int) → set[int]: IDs of the nodes connected to the given one by an edge (in any direction)

getNodeDistances(source: int) → dict[int, float]: Topological distance (number of edges) from source to every reachable node

getNodeIDs() → list[int]: Return a list of all node IDs

getNodePose(node_id: int) → mrpt.poses.CPose3D: Get the estimated pose for a node

hasNode(node_id: int) → bool: True if a node with the given ID exists

insertEdge(from_id: int, to_id: int, edge: mrpt.poses.CPose3DPDFGaussianInf) → None: Insert a directed edge from → to

loadFromTextFile(fileName: str, collapse_dup_edges: bool = True) → None: Loads the graph from a text file in the TORO / g2o format.

nodeCount() → int: Number of nodes in the graph

saveToTextFile(fileName: str) → None: Saves the graph to a text file in the TORO / g2o format.

setNodePose(node_id: int, pose: mrpt.poses.CPose3D) → None: Set the estimated pose for a node

mrpt.graphs

Package Contents

`mrpt.graphs`