From the description it sounds like the waymo did everything right and the cyclist came out of a blind spot.
Have seen this exact scenario play out many times and it doesn’t end well for the cyclist.
Would be better for all parties if drivers were more aware of bikes, but I think that has gotten worse over the years, especially with people who can’t stay off their phone while driving. It’s scary out there.
I think I read somewhere that the entrance to the intersection had a regular lane and a bicycle lane so it may be that the bicycle entered the intersection at the same time as the truck and was slower than the truck through the intersection.
That would be a little more concerning as a legal and reasonably predicable traffic setup as opposed to hitting somebody who was out of sight until the last moment because they are in an illegal situation.
Still would expect this to happen to many humans in the same situation. Honestly, the detail that the vehicle braked aggressively seems positive.
Disagree… it’s not right to not see a cyclist.
My husband, a cyclist, jokes that his bike is his invisibility field. It’s actually really hard to see a bicycle. Computers may ultimately do better than humans, as the movement a bicycle makes isn’t really the kind of movement we are programmed to see. (It’s too regular.) Bicycles popping out from behind trucks are are especially hard to see.
That the car brakes hard means the car DID see the cyclist, just too late.
I think the car could be anywhere from not-at-all at fault to mostly at fault, depending on details we don’t know.
Fixed that. Though it might be implied, I prefer that engineers get their explicit due, credit or blame.
Not saying it’s right, just saying it’s very common for drivers to not see cyclists, especially if the cyclist is in a blind spot, like behind a truck.
The fact that the waymo did brake means it did see the cyclist, just too late. I’m figuring that a human driver in the same situation may have fared better or worse, depending on the conditions. We don’t have enough information to know for sure.
Was stuck behind a garbage truck today until a suitable break in oncoming traffic for me to go around the garbage truck by crossing the center line and going completely into the oncoming lane.
Does anyone know where they are with testing stuff like that on autonomous vehicles? Will self-driving cars go around mail / garbage trucks, or will they just hobble along at an average speed of maybe 0.5 MPH until either they or the truck makes a turn???
At this point they aren’t really testing, they are doing. Cruise and Waymo did millions city miles last year, so they’ve probably dealt with every common thing like that.
But I don’t know what they do. How much emphasis they have on following the rules. And when they break a major law, whether it’s automatic or remote or both. Might ask Reddit.
Waymo is now cranking out around 2M miles a month, which would make me nervous if I were an engineer.
So far Waymo isn’t doing full self driving in my area. They are just collecting data and testing. I’ve seen another Waymo car in my neighborhood since this pic, and my wife saw another one as well.
For some reason I initially read that as colliding.
Column A, column B…
The answers seem quite insufficient. “Yes but only in a few years” covers a breadth of assumptions. What about “Yes, only if they’re proven to be safer than human drivers” or “Yes, already today, but only for certain predictable uses like long-haul trucking”? “No, never” seems like an extreme position that’s forced in as one of only 3 options. No option for “Yes, eventually, but not in the next few years”?
Yeah, the no/never is a ridiculous option. I would have preferred it to say Not For Many Years.
“No, never” is about as never as the next survey, in a few years.
People can change their minds when auto-autos ™ improve.
I don’t think they ever will, but what do I know?
Not exactly a response to your post, but related, and more insane.
…In the last example, Waymo swerves into oncoming traffic to avoid a crash.
Yeah, that’s not addressing the concern I raised.
