Validated speed accuracy over a fixed distance
To accurately assess the absolute speed and distance measurement accuracy of the DL2, we set up a laser trap system as shown in the diagram below. A fast laser sensor was mounted in the car and driven past two reflective markers at the side of the track approximately 40m apart. The DL2 was set to log when the vehicle passed the reflective markers as well as the GPS data. The laser detector pulse was time stamped with a resolution of 0.16μS.
A total of seven passes were made; the first five were made at a relatively constant speed, for the final two tests the vehicle was driven into the trap at a constant speed and then maximum braking was applied and maintained through the second trap. In the past this has been a test which has showed serious problems with GPS based speed/distance measuring systems as any latency in the measurement or any filtering in the speed measurements causes large errors in the computed distance.
From the results it can be seen that in the case of the 20Hz DL2 system, the maximum discrepancy between the measurements is 6cm, and in the case of the 5Hz system this increases to 17cm.
Of the 6cm error from the DL2, a large proportion of the error is very likely to be experimental rather than true inaccuracies in the system – for example a small course change of the vehicle would account for a significant proportion of the error, or even very small angular movement in the laser detector would account for the errors seen. In conclusion, the results suggest that the DL2 GPS system is free from any significant time misalignment, and measured distance accuracies are within a few centimetres over a typical braking/acceleration test.
If we assume the limiting case and all the measured error is from the DL2s measurements, rather than from the experiment then the equivalent average speed errors are: