4 way stop signs were really confusing for me when I first learned how to drive. It seems there's so many nuances about when you should go or not. Is there even an algorithm available to handle them?
According to a NYT time article [1] they can already do that. I'm sure by this point they can model other vehicles behavior in common situations like this. But ktRolster brings up a good point, there are many situations where a glance is all that is used or needed to signal intent. In the article they mention that at four-way stops they do this by just inching forward and testing the waters, but obviously you don't want to inch forward into a pedestrian or bicycler. Makes me wonder if we might want to standardize the method of communicating intentions to pedestrians across all self-driving cars.
