Interpreting Raw Tracks
Raw tracks form the source in which we build and improve all our maps. The raw tracks actually contain a lot of information, useful to help a mapper understand the circumstances under which the tracks were recorded.
Accessing track properties
The 2 common ways of accessing Track Properties:
- Via Mapsource - For tracks which are saved in .gdb format - Open the tracks in Mapsource -> Tracks Tab -> Select a track -> right click -> Track Properties.
- Via MapEdit - Highlight a track which has been added to your map -> Right click -> Properties -> Points tab.
Information from track properties
Information that can be obtained from Track Properties (using the Mapsource Track Properties as an example. The Mapedit version is very similar except for header labels):
Time and Date
This shows the exact time and date that each coordinate was logged.
Distance travelled between consecutive points logged. The shorter the distance, the more accurate the track will be.
Time interval between consecutive points logged. The shorter the time interval, the more accurate the track. If 2 tracks of the same road are taken, 1 with Leg Time of 1 sec and the other with time interval of >1 sec, we would place greater reliance on the tracks taken at 1 sec intervals.
Road curves on 1 sec tracks can generally be fully relied upon to reflect actual road conditions as sufficient points are logged while negotiating the curve. Road curves on tracks which have leg times between 5-15 secs are generally unreliable. This is because points are logged too far apart resulting in "missing" points while negotiating the curve. Instead of forming a nice curve, the tracks tend to represent curves as "curved straight lines". Need to add additional points while drawing the polyline to better represent the curve.
Shows the speed at which the tracker was travelling when the points were logged. In residential areas, this gives clues as to whether the tracker recorded the tracks in a car, on a bicycle or by walking. This is important in determining the final position of the polyline vis-a-vis the tracks.
If the tracks show speeds of around 20km/h +/-, this is usually taken in a car. As the tracker is likely to put the GPS or antenna on the driver's side, it follows that the tracks are likely to be closer towards the center of the road. If the tracks show speeds of less than 5km/h, it is likely that the tracks were either taken using a bicycle or while walking. In this case, the tracks are likely to be to the left edge of the road - the actual polyline should thus be drawn to the right of the track to compensate for this.
This shows the direction in which the tracker was travelling and thus, tells us which side of the road the tracker was travelling on. The polyline should ideally be drawn slightly to the RIGHT of the track based on the direction in which the tracker was travelling. This is to compensate for the fact that the tracker never actually drives exactly in the middle of the road.
This shows the exact Lat/Lon of each point logged.
There is no specific field in the tracks showing the accuracy or Estimated Positional Error (EPE) given on the GPSr. However if examine regular 1 secs or 2 secs tracks, we may see missing track points. These missing points indicates where the GPSr didn't manage to record any position, in other words the satellite reception was lost at those points, hence the preceding few points are suspect in terms of accuracy.
If the tracker used GPSProxy to record the track logs, the original PLT file does include the EPE reported by the GPSr at each point location.
Information other than from the tracks
Lastly, it is important to know whether the tracker was using an external antenna or not when interpreting tracks. You will need to get this information from the tracker - this does not show up in the Track Properties. Tracks taken with an external antenna are usually "smooth" looking and consistent ie. straight road stretches show up as straight lines. Tracks which were taken without an external antenna are generally jagged, inconsistent. Care needs to be taken in interpreting the actual road alignment. Generally speaking, tracks taken with an external antenna are far more reliable than those taken without an external antenna.
The best way to understand how to interpret tracks is to take your own tracks of the SAME area. Drive on the same route 2 times, in the same direction, the first time with your track log set to 1 sec intervals, another time with your track log set to "Auto, Most Often". Make sure you make a few turns to see the impact of the track log settings in curves.