Ask HN: Why are we building self driving cars to navigate based off human signs

7 points·vincefutr23··6 comments

Instead of building self driving cars to visually read stop signs and see other cars with computer vision, why don't we put some sort of chip in signs, lane markers etc that are inexpensive and designed to be read by machines? Would cut down on the difficulty of the problem in my opinion and increase safety

View on HN

"Inexpensive" is really an understatement when talking about replacing the entire, worldwide sign and lane marker infrastructure (nb with something that's orders of magnitude more issue-prone than current solution). Who pays for that? What if a state/country says "we don't care about your active signs, WONTFIX"?

Worse, "can't see stop sign" is currently easy to spot and debug; in your proposal, this would require retooling the world over.

1958 "self driving" car: https://www.youtube.com/watch?v=xKOdux6Gjno&feature=share - There is really nothing stoping someone from taking a different approach to driverless cars. Smart highways is not a bad idea at all. In the same way we have a carpool lane, one could provide a driverless interstate lane that is faster and more efficient by equipping this lane with a guiding system. #hyperlane

A guiding system? Might we say that such a system is "on rails?"

ducks

Interstate driving is already workable. It's city and rural traffic that's hard.

What happens when you come across a sign that doesn't have a special chip?

How many stop signs are there in the world? That almost sounds like a Fermi problem...

What if a sign or chip gets removed/knocked over . . . or swapped out maliciously . . . sounds more dangerous.

Reply on news.ycombinator.com