Hi all, I would like to know if for Open Street Map there is a limit with which the separation accuracy of objects must be guaranteed on the map. This is the relative positional accuracy that differs from the absolute accuracy because it does not deal with the position of objects in space. I am not interested in this relative accuracy value for large areas, but for small areas for example 4-5 square kilometers. Thanks for your attention asked 16 Apr '20, 18:00 SergioV |
Generally in OSM, there is few rules, mostly guidelines, do as best as you can... I'm not sure I really understand your question, but usually the quality of objects' positions in the database is incremental. Sometime we start from crude GPS traces, then refine with blurred imagery and memory, or even street level photo and some day we get crystal clear recent imagery, and we redraw, or move, everything in its place. We try to be as precise as possible, knowing that most of our tools are not precise at all (consumer GPS is around 5m in good conditions, for example). The data consumers need to be aware of that. Regards. answered 16 Apr '20, 22:37 H_mlet Thanks for the answer, I understand what you're saying, but my question was about cities where the maps are well established. In these cases and, I repeat for limited areas, what can be the accuracy of the measurement of distances between buildings or the orientation of the streets? There may be errors in the size of buildings, but they seem to me to be isolated cases. I tried to make some comparisons between OpenStreetMap, Google Maps and the GPS sensor in RTK mode which is much more accurate than normal GPS as it can reach an accuracy of a few centimeters. The tests were conducted in Turin and in a neighboring country and I observed that the distances between the same points of the two maps are within half a meter, that the orientations of the same roads are within a few tenths of a degree, that the GPS position is within 1 m from that indicated on the map. These are values that I was able to measure experimentally and I wondered if it was possible to establish a valid value for cities or countries that have a consolidated map based on the processes and sensors that are used to create the maps. I also wonder why the GPS in RTK mode, which I mentioned and which is very accurate, despite being a low cost, I can not find it among the sensors mentioned in the forum questions
(17 Apr '20, 16:37)
SergioV
1
I had to look up RTK it's here https://en.wikipedia.org/wiki/Real-time_kinematic
(17 Apr '20, 23:19)
andy mackey
|