This project aims to find a way to plan trips based on my (or your own) historical geolocation data. Several companies collect your location, the best known fo this is Google, just go to (Google Takeout)[takeout.google.com].
I'm currently trying to find a way to find paths in Budapest.
If you have locaion data from Budapest, you may be able to follow my scipts.
To start, clone this repo, and create data/raw and data/edited folder in the project root folder, then copy your Google location history data into data/raw.
basic_bp_map.Rfilters inner Budapest data, creates a dataset accordingly.get_distances.Rfind out the longiude and latitude change per meter, then tries to naively find the two endpoint in a 2-hour range. You can skip this script.create_routes.Rlooks for continuous recordings in time, and labelling them - at the end creating a new dataset again.trip_intersection.Rattempts to find spatial intersection between routes.
If I can find intersections properly, then I will have to find paths on based on spatial data. If that works, I can find how long each path segment takes from one intersection to another. In case of multiple paths, I can find alternatives, and based on speed, I can infer travel mode.