Closed kozyr closed 7 years ago
You have quite large deviations of GPS positions, probably because of the buildings. You could try increasing the matcher.sigma value to, e.g., 8 meters.
Tried it with 10 meters as matcher.sigma, but it still picks Kurze Muhren...
Could you please provide the raw data in the format JSON format required by Barefoot?
Sure. The point that goes to Kurze Muhren is with datetime of 1475684525936
Okay, I just checked back and got a path without Kurze Mühren if sigma is set to 15 meters, but I think some other matches will fail then especially because this is in a dense city area. Instead of increasing sigma to some large value I would try to increase the data quality, e.g., with heading information, or data quantity, e.g., higher frequency of position measurements.
Ok, thanks. This data though comes from normal consumer phones, and it's not practical to sample them more often than once every couple of seconds. I could get the heading information off the phone, but not sure whether that would help here..
Are the users walking or going by car. If by car, heading will help.
I'm closing this issue due to inactivity, please reopen it if there is more on that.
Hi, I'm looking to see if I can use barefoot for my projects, and run into this in my first test. My raw data is as follows
and when I match, I get points onto Kurze Muhren (point 27), which was actually never driven on. Is there a way to match point 27 onto some other road (Bransende in this case)? I'm running everything right now with default parameters.