Track snap to Map: Matching GIS algorithm (.gpx G P S track to mapped roads)

Import G P S track (as a .gpx file - example attached). This compares the track to the map, and map matches to deduce the actual road segments likely to have been used. The output is another .gpx (or .kml) file, containing ONLY the matched road segments. Project can be .exe code, Java, web service or any technology which achieves the task well, and is open to later extension. Road map data to be used should be [[login to view URL]][1], and exported track to contain ID references of the Open street map segments which were matched, and closest matching time stamps on matched track from original. The matching algorithm must provide an assumed logical path, i.e. it is trivial to simply match each point to the closest road segment, but this gives disjointed tracks. Rather, the matched path must form a 'connected route' which a car may have driven. Matching off-road (data) movement is not required. On completion of matching, both tracks to be shown on either Google Maps or Open Street map, so comparisons can be made. Please feel free to ask me any question on this project, prior to bidding. ## Deliverables The use of the <[login to view URL]'s_algorithm> or equivalent may be used to assist matching, given the vast majority of routes follow optimum paths from A to B. Similarly there are many map matching algorithm examples online such as: <[login to view URL]> <[login to view URL]> <[login to view URL]> <[login to view URL]>