|
A Tesla driver reports the following issue:
His route was
Tesla is rumored to be using OSM (via MapBox and Valhalla) for in-car routing since April 2018. Where can I go (other than into a Tesla) to simulate this route knowing that it will be based on OSM/MapBox/Valhalla? I tried https://www.tesla.com/trips, but it is apparently based on Google, and the routing was precise. The Dave & Busters the user wanted to go to is mapped on Google. It is not mapped on OSM. I am trying to construct (and demonstarte) a scenario, of which OSM is part, where a POI does not exist, yet it is somehow found (from an external POI database? by address? street number extrapolation?) and is mapped to the GPS coordinates but without regard for the elevation. |
|
You can click the doubled arrow button on the upper left of https://www.openstreetmap.org/ to experiment with routers using OSM data. I don't know of a public Valhalla instance. Thanks, can't reproduce the issue navigating by endpoint address, because, naturally, it leads to the correct street. You answered my original question, but the real question for me is, how can the described issue be possible? Since the POI he wanted is not on OSM, it had to come from a third party. I doubt most POIs have altitude coordinate. If the lat and lon placed it under the freeway, then the nav software, whether using OSM or not, would be tempted to navigate to the largest nearest road, i.e., the freeway. Does that sound about right? I don't see how OSM can be blamed -- or for that matter, used to fix the issue. Other than by adding the POI. And that's IF Tesla is really using OSM, which is a big "if".
(25 May '18, 01:39)
ponzu
|
