Using a stereo panoramic interactive navigation system to measure pedestrian activity scheduling behaviour: a test of validity