Related
RefineTable of content
This refinement keeps a single connected edge for each point for each point: the one with the highest score based on connected 
🝰 Heuristics.
Note that the remaining
Edgecan be the same for multiple, differentVtx.
Not all heuristics can be yield usable scores outside of a search context. For example,
Available Heuristics Modules
🝰 Heuristic Attribute
Attribute-driven heuristics
The Attribute heuristics uses custom point or edge value as raw score.
🝰 Feedback
Favor uncharted points & edges.
The Feedback heuristic add/remove score value to points & edges that are “in use” by other previously computed paths.
🝰 Inertia
Favor active direction preservation.
The Inertia heuristic uses the ongoing traversal data to try and maintain a consistent direction, as if the algorithm had “inertia”.
🝰 Steepness
Favor flat trajectories.
The Steepness heuristic uses the edge angle against an up vector to compute a dot product that is used to determine whether the edge should be considered flat or not.
🝰 Azimuth
Favor edges directed toward the goal.
The Azimuth heuristic attempt to force the path to always aim toward the goal.
🝰 Least Nodes
Favor traversing the least amount of nodes.
The Least Nodes heuristic favor node count traversal over anything else.