I don't have much background knowledge in mapping or self-driving, so I thought this was well-written in a way that anyone could understand! Re: breaking down the data framework into world geometry vs. semantic annotations— that makes sense, but how often would you need to update the annotations? Just running and biking around San Francisco, I feel like there's no shortage of construction that create changes to lanes/stop signs/lights and rules of traffic flow. How do you account for these, or would world geometry data be enough here?
Is there a way to know the "freshness" of the data? Obviously if I'm driving and a car in front of me just imaged the road I can be much more reasonably certain that there isn't any debris on the road surface, or new potholes that have opened up.
Hi all! I'm Greg and I work on the Mapping Team at Aurora. I'd be happy to answer any questions about the article and mapping for self-driving.