Monday, March 16, 2009

On the road

Performing real-world experiments often takes significantly more time than just thinking about them from the ivory tower. At the same time the amount and things you learn can be great, it's not just the results of your experiments but also the process of setting them up.

This weekend we took one of the CVIS systems for a test drive to collect data for a demonstrator we're developing. Besides the data we got, which will really help to test our code, we also learned (or were reminded of) the following:
  • Turning off the ignition after half an hour of data recording will cut the power from the 12V socket, effectively messing up your measurements (yes we should get some form of UPS)
  • Turn on compression when recording video since the recording software doesn't handle files larger than 2GB
  • People at bus stops start to stare the 10th time you drive by, stopping for invisible pedestrians.
We used two neodymium magnets to attach the antenna to the roof and a basic webcam to get some context for our recordings.