I am new to the GIS-World and just learned a couple of tools. Now I want to start with my actual project, but since i don't have a overall understanding I would have to hear your opinions about the feasability of my ideas.

I have a lot of data points which refer to car drives getting tracked by GPS. I want to group points which are really close to each other to be able to compare their sensor values.

First I thought of rounding their values to something around 5 decimal places to group them, but then the points would not lie on the street anymore.

My actual plan is to load german street data from OpenStreetMap-Data using osm2pgsql into PostGIS. Then I want to calculate the distance of my data points to the closest street using somehow (maybe ST_Distance). Then I would like to find some algorithm to group points on the street by some flexibel interval (e.g. 5 meters).

Do you think this might work? Any other tools or data I should have a look for? Just trying to minimize programming time invest into some system which might not be able to work.


Well, I don't really like to answer own posts, but I just found another gis stackexchange post which lists ideas and tools for "map-matching" which seems to be the agreed term for this kind of tasks. Propably there you will find answers if this question is meaningfull to you.

  • 1
    Always great if people can answer their own questions :) – Joseph Oct 6 '15 at 14:05

I would suggest setting up a topology and setting a snap tolerance for your streets and points.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.