Our great sponsors
-
road_surface_classifier
Leverage machine learning to aid in the tagging of drivable ways (i.e. roads) in OpenStreetMap.
-
InfluxDB
Power Real-Time Data Analytics at Scale. Get real-time insights from all types of time series data with InfluxDB. Ingest, query, and analyze billions of data points in real-time with unbounded cardinality.
Komoot uses OpenStreetMap data. That is a huge plus in the many countries where their data is much better than the likes of Google, but if youbhappen to be somewhere the data isn't great, then you can get results like that. The good thing is though, that if a path is illegal to cycle on, all it takes is making an account and adding bicycle=no to the path. Your edit is applied immediately (though might be reviewed by a human later) and will result in better data in Komoot but also in other apps like Strava or cool cycle infrastructure maps like www.cyclosm.org
Someone has been working on a very similar project and published their source code. See https://github.com/jdalrym2/road_surface_classifier - there's a thread about it on https://community.openstreetmap.org/ . He did a talk at the State of the Map US conference just gone (videos aren't up yet but should be in a couple of weeks).