On Sat, May 1, 2010 at 9:03 AM, Dave F. <[email protected]> wrote:
> The route should be *accuratized* (Is that a word?). > Change the route to where you actually went, not where your inaccurate > GPS showed. > > Simplifying makes it less inaccurate. I fail to see how replacing measured data (with a known variability of accuracy) with a guess (with a totally unknown variability of accuracy) will yield a superior result. If the GPS is assumed to be inaccurate, and guessed locations to be more accurate, then why bother buying a GPS just get after mapping the world by randomly plunking down nodes, and drawing ways between them. A good quality consumer grade GPS with a clear view of the sky will give fairly accurate results. Rarely have I ever gathered information that would put me outside of the width of the pavement while gathering data on a roadway. On the other hand, in areas with with degraded signal coverage, the GPS can give erroneous results, but not because the GPS is inaccurate, but rather because of the multipath RF reflections, or just plain low signal levels from the GPS constellation. I agree that GPS traces need to be vetted for anomolies, and possibly sections removed where the recorded track may backtrack or overlap itself due to the path chosen by the operator. Referencing other material such as Yahoo imagery at this point becomes very helpful, but even there, the Yahoo imagery can have an unknown offset value. While our consumer grade GPS receivers can't yet give us sub-centimetre accuracy, given the opportunity to have a good view of the sky, then can provide some fairly accurate results, much more so in my opinion than simply guessing. James VE6SRV _______________________________________________ newbies mailing list [email protected] http://lists.openstreetmap.org/listinfo/newbies
