It was not a good first day on the job for the driverless shuttle.
I’m amused by the futile attempt to spin this story. That it performed as designed. Yeah, right. That turd stuck on on the bottom of your shoe doesn’t stink either. :rolleyes:
That’s b.s. I’ve been a professional programmer/ Analyst for twenty plus years. Forgetting to implement a key design element is not success. Sometimes the programming team has to regroup and meet with the departments that use our software. Retool the design to fully meet their expectations and evolving requirements. The departments (requesting the new software) don’t always know what’s needed when a new project is being discussed and developed.
The shuttle programmers forgot two key elements. A loud horn and the ability to take evasive action. D’oh! There is some confusion whether the car’s programming has even limited abilities to evade an accident.
I’ll pass on taking this shuttle. I prefer walking. Thank you anyhow, Vegas. LOL
The way I see it, this absolutely demonstrated that both driverless cars and human drivers can be very flawed.
I think driverless cars are not yet ready for prime time - but if they can reach a certain level (and I think they can, one day), they will be far better than human-driven cars. I’m also a proponent of going directly to level 4, and not having the “it can drive by itself part of the time” cars on the road because the humans who are supposed to know when they can’t drive on their own will be stupid and no do what they should. Also hate Tesla’s “auto-pilot” branding and find it irresponsible.
A loud horn is nice. But plenty of human-driven accidents happen with horns blaring all the way to impact and the other driver still clueless about the collision they’re about to cause.
One of the key factors about AVs is they will have different types of accidents than humans will. They will avoid umpteen screw-ups we make and make a smaller number of screw-ups of their own.
Which means that the majority of AV accidents will have an element of “Duh, every human driver knows not to do that! Stoopid compyooterz!”
This factor will be a big issue for the law, the manufacturers, and the PR. Humans are real good at group ridicule. A few early surprises like this can get the whole project labeled as hopeless.
The fact we treat tens of thousands of human caused accidents as mere background noise and will be both publicizing and investigating the shit out of each and every AV accident makes the problem even worse. Imagine every single human-driven traffic accident had a team of experts interviewing witnesses, reconstructing wreckage, and poring over data recorders and vid surveillance footage for a month or two before issuing a public finding.
The driverless shuttle got into a fender fender bender the first day on the job. That is an accurate statement.
You are hired to run errands. First day on the job, you are sent to pick up supplies at a store across town. You come back with a dented company car. Your boss will be pissed. You can say it was the other guy’s fault and the police report supports your story. You won’t get fired.
You still got into an accident. Defensive driving is expected from all of us.
indeed, I like to bust out this quote from the Joker in The Dark Knight, since it sums it up pretty well:
“You know what I noticed? Nobody panics when things go according to plan. Even when the plan is horrifying. If tomorrow I told the press that, like, a gang-banger would get shot, or a truckload of soldiers will be blown up, nobody panics. Because it’s all part of the plan. But when I say that one little old mayor will die, well then everybody loses their minds!”
Too bad. I hope they fix the glitches and get it back on the road. I’d even consider going to Las Vegas, which isn’t my favorite place, just to ride it.
By the middle of the next decade, we won’t need any mass transit because everyone will be summoning driverless taxis on their cellphone apps, or so says an op-ed in today’s paper.
I’ll be home enjoying the reports of mass collisions and chaos.
The problems will get worse before they get better. I don’t know how these live road tests are allowed, if I set my car running down the street without a driver in it I’m sure I’d be held responsible for any accidents no matter whose fault it was, it would probably be considered a crime even if no accident resulted. Somehow people have been blinded by the allure of technology and believe we can loose driverless cars out on the road because someone claims they have programmed a computer to drive. We need this technology to be developed, but the current approach of putting these faulty prototypes on the road is going to create huge problems.
A little bit of irony here is that the accident probably wouldn’t have occurred had the truck been a driverless car with the proper sensors to notice an unmoving obstacle behind it…
It frustrates me that commercial vehicles don’t have the more modern sensors.
I bought my Econoline E250 Cargo Van in 2013. Ford offered no driver assist options. Zero. The only option was a backup camera in my mirror. I did buy that.
I would have gladly purchased drivers assist. Especially the warning sensors that I’m too close to another vehicle and collision avoidance braking.
2013 was the last year for the Econoline Cargo vans. I had to scramble to get the funds to buy one before they were gone.
They were phased out for the European Transit vans.
When all the vehicles are driverless, and communicating with each other, it will all work much better. I imagine at some time in the future cars with drivers will be restricted from operating in certain times and places unless they can be placed in automatic mode. On the way to that point in time driverless cars should be restricted from operating at all times and places where cars with drivers are being operated unless they can be operated in manual mode by a driver. There should be a huge test facility somewhere automatically and manually driven cars can simulate real world situations and create a more perfect system before the public is exposed to the risk. The current standard where a bunch of programmers, managers, and marketers say “Hey! We think we’re ready to put this on the road” is way too low. It’s way too low a standard for technology that isn’t likely to kill people already, and we’ve already had exploding batteries and GPS that directs drivers off a cliff because of the low standards of acceptance we place on technology. Eventually driverless technology can be demonstrated to be reasonably safe for use on the public roads by some objective standard, but there’s no real way to measure or demonstrate that now without placing people in danger.
On more than one occasion, I’ve been a passenger in a car that was in danger of being backed into. The driver just sat and stared (apparently in disbelief–I don’t really know) and did absolutely nothing to try to avoid the collision.
From the photo in the article, I’d say that one contributor to the accident was that the shuttle is small. The semi driver who decided to back up is definitely at fault, but if it had been bigger, it might not have been overlooked.
Isn’t the general PR line that a possible driver will need to be present in any driverless car? And isn’t the only thing the boxed-in car could have done is honk the horn? Wouldn’t the human driver/passenger have done that in an actual, non-research, situation?
The comments here make me think that there will need to be a public outreach campaign warning people that there are going to be more cars on the road that follow the traffic laws instead of, say, speeding up to try to get into the intersection as the light is turning red.
According to a tweet from the company operating the shuttle: “Shuttle is capable of reversing directions. It didn’t in this case, because cars were stopped behind it. It was essentially boxed in, says Chris Barker, VP of new mobility at Keolis.”
