[05]

> They don’t drive like humans, they don’t negotiate intersections like humans, and they’re going to interact with human drivers and pedestrians in unexpected ways.

Well, if you can't make a car that behaves like a good human driver, then maybe you shouldn't be making driverless cars? Otherwise, instead of reducing accidents, you might increase them (at least initially, before humans adapt). Self driving cars that can't interoperate with humans using existing infrastructure are basically useless - infrastructure lifetimes are orders of magnitude longer than technology life cycles. Re-planning cities for driverless cars now will make you regret it when the technology matures.

Also, while centralized planning might fly in current day '1984' London, I sure hope this fails at least in EU, where they claim to value privacy. Nothing about self driving cars requires either centralized reporting or even V2V - if you can detect a pedestrian behind a tree, you sure as hell can measure the speed of the car ahead.

[mc32]

RE: V2V, I'd like to see that car ahead of mine tell mine "hey, from my pov, there are people approaching, you'll detect them in five seconds --slow down as precaution. Same thing for things like black ice or object in the middle of the freeway.

[toomuchtodo]

Do you need that level of complexity if you simply slow down because the vehicle in front of you that detected pedestrians slows down?

[05]

Yeah, the way I see it it's a convenience feature at best - if you car can't brake in time for the people around the corner, it has no business driving the current speed. In the end, you can't rely on other cars telling you it's safe - that's bad engineering.

There are second order benefits, like slowing down with less acceleration, so your coffee won't spill, but compared to the infrastructure investment and privacy issues, those things are really negligible.

[viraptor]

There are many cases where this could be useful. Random one: let's say you're on a motorway behind another selfdriving car. It can see something: not moving, on the road, in the distance. Your car can't because the view is blocked. If the car in front is starting to slow down, then you've got two options - slow down as well, or start overtaking.

Without the extra information, it's a completely reasonable decision to start overtaking. And then you slow down as well, after spotting the same information. So yes - V2V would help here.

[05]

> Without the extra information, it's a completely reasonable decision to start overtaking.

Only reasonable if 'defensive driving' is a curse word in your dictionary. Otherwise it makes sense to wait a bit and see if you can detect something too. BTW, Google reported radar working 'through' at least one car, it can look through the gap between the car and the road surface. And LiDAR can see through glass of a car in front, or just over its top (the sensor bump of the car in front down' cover much of you car's view when cars are separated by several car lengths).

[viraptor]

The technical word is threshold. You're saying "wait a bit" - sure. Now you waited a bit and still don't see anything - you're going to start overtaking.

[mc32]

No, rather it illustrates advantages of v2v comms. It could also aid merging to make it more seamless. Or, in cases of sudden stops or roadside emergencies, the car at the vanguard could alert the other cars trailing it of any upcoming driving conditions of interest.

[coldtea]

Sounds like a hack waiting to happen...

[AznHisoka]

What if there is no car in front of you?

[toomuchtodo]

Then you detect pedestrians as you normally would. I'm stating the additional vehicle to vehicle communication regarding pedestrians is overly complex and unnecessary.

[coldtea]

Then nobody covers your field of view and you can see those pedestrians immediately and directly?