Today, I noticed something. It has nothing to do with GPSbabel, but I
thought you might be interrested. So I share it with you, readers.
A .tcx-file is the output of a Garmin Forerunner sportswatch. It is a
gpx/xml like file. Every trackpoint has a number of data-fields. One of
them is the "DistanceMeters", the total cumulative travelled distance
since the unit was reset.
When doing some track analysis, I noticed that the this distance can
actually decrease for successive points in time. As if the GPS would
notice when I would be running backwards, momentarily.
I guess it has to do with the internal filtering. The track data is
written every second to the unit's memory. The low-pass position filter
in the unit may correct the position with some delay. That means that
with the knowledge at time t, the position at time t-1s or t-10s may be
corrected. And that means that for example an overshoot could have
occurred. Apparently, that is my guess, the unit corrects the total
cumulative distance by placing the current point somewhat back on the
track. Not assuming the runner actually runs backwards, but just using
the best guess knowledge. This might improve the accuracy of the total
running distance, but it is strange when you look at the raw data.
Jan-Gerard van der Toorn
Get latest updates about Open Source Projects, Conferences and News.