This is, as the name suggests, yet another map matcher (see this page) between a series of observations (mainly noisy GPS points) and the most probable path on a given graph.
It uses a first–order Hidden Markov Model, Viterbi algorithm, A* search algorithm, Ramer–Douglas–Peuker algorithm for point simplification, Bentley–Ottmann algorithm for intersection between points and polylines, Graham scan for convex hull calculation, Kalman filter for smoothing the observations, HPR tree to get the list of roads, various geodetics algorithms, and a graph merger builder that connects also near edges.
| This project adheres to the Zero Bugs Commitment. |
|---|
Please report issues to the issue tracker if you have any difficulties using this module, found a bug, or request a new feature.
Pull requests are welcomed.
- First version.
This project is licensed under MIT license. For the full text of the license, see the LICENSE file.