Ex-Uber Engineer Claims a Self-Driving Car Drove Him Coast-To-Coast (theguardian.com) 114
"Anthony Levandowski, the controversial engineer at the heart of a lawsuit between Uber and Waymo, claims to have built an automated car that drove from San Francisco to New York without any human intervention," reports the Guardian. Levandowski told the Guardian that he completed the 3,099-mile journey on October 30th using a modified Toyota Prius, which "used only video cameras, computers and basic digital maps." From the report: Levandowski told the Guardian that, although he was sitting in the driver's seat the entire time, he did not touch the steering wheels or pedals, aside from planned stops to rest and refuel. "If there was nobody in the car, it would have worked," he said. If true, this would be the longest recorded road journey of an autonomous vehicle without a human having to take control. Elon Musk has repeatedly promised, and repeatedly delayed, one of his Tesla cars making a similar journey. A time-lapse video of the drive, released to coincide with the launch of Levandowski's latest startup, Pronto.AI, did not immediately reveal anything to contradict his claim. But Levandowski has little store of trust on which to draw.
So What (Score:1)
I still want my flying car!
Re: (Score:1)
Get a Tesla and find a highway off ramp with a drop-off instead of a concrete wall.
Re: (Score:3)
I still want my flying car!
Then go buy one, they DO exist, although I seriously doubt you can afford to purchase it and keep it airworthy. There have been a number of designs, some actually built and flown. Come to think of it, you might have better luck building your own experimental aircraft/car and it might be something you could actually afford to kill yourself in.
Flying isn't hard under ideal conditions, if I can do it, almost anybody can. However, knowing how to stay out of trouble when conditions are not so nice or when some
Re: (Score:2)
I was an aircraft mechanic for 45 years. ;)
I don't want to do any work on it, I don't even want to fly it. . . .
I just want the flying car I was told we would all have by now
Re: (Score:2)
There has never been a flying car, and there probably never will be. There are only roadable airplanes. A flying car wouldn't require a preflight check, and it would be able to take off from your driveway, or the interstate. Nobody has ever built such a vehicle, and limitations of physics suggest that they never will — not because we'll never get enough energy into the vehicle, but because getting it out would crack pavement, flip over neighboring vehicles, roast pedestrians, etc.
Re: (Score:2)
There has never been a flying car, and there probably never will be. There are only roadable airplanes. A flying car wouldn't require a preflight check, and it would be able to take off from your driveway, or the interstate. Nobody has ever built such a vehicle, and limitations of physics suggest that they never will — not because we'll never get enough energy into the vehicle, but because getting it out would crack pavement, flip over neighboring vehicles, roast pedestrians, etc.
Oh, right... Um, If you let me define a "flying car" anyway I choose, I could prove my point too.
https://en.wikipedia.org/wiki/Aerocar
It flew as well as drove on roads. No airworthy and road legal examples are for sale, but they do exist, or, if you wanted to license the plans you could build your own and fly it as an experimental aircraft and drive it as a home built car in some states.
Your "I don't want to have to do the conversion or tow the wings in a trailer" and the "Operate off a road" are but re
Re: (Score:2)
"Your "I don't want to have to do the conversion or tow the wings in a trailer" and the "Operate off a road" are but regulatory restrictions."
Completely false. Those are physics considerations.
" But the fact remains, there are vehicles that can fly in the air on their own or drive on roads on their own, and they've been around since the 40's."
And the fact remains that you have to stop and do a bunch of stuff to them before you can fly, in fact literally converting them to airplanes before you can fly them.
Re: (Score:2)
I still want my flying car!
Neil deGrasse Tyson explained why we don't have them yet.
https://twitter.com/neiltyson/... [twitter.com]
Sometimes I wonder if we'd have flying cars by now had civilization spent a little less brain energy contemplating Football.
Color me skeptical (Score:4, Insightful)
This guy is mostly famous for being a big liar and a thief. Not buying it. Also not sure why anyone would care about this.
Re: (Score:2)
Even so there may be things like road works, lane closures, accidents, weather, speed restrictions, tolls etc. on roads for which no automated car could reliably cope with. And since there is no one single continuous road across the entire continent, there are also
Re:Color me skeptical (Score:4, Insightful)
This claim contains an obvious lie. The car does not have 3000 miles range. It can't refuel itself. Even if it could, it can't clean all its cameras if they get dirty (Tesla has the same problem). And one illegal run doesn't prove general capability anyway.
Re: (Score:1)
He did stop to rest and refuel. He could have cleaned the cameras at the stop. Seems very plausible. Oh wait, no it doesn't, because "autonomous driving" is a joke.
Re: (Score:2)
That was my point. He said it could have done it without a passenger, but seemingly not.
Re:Color me skeptical (Score:5, Funny)
Re: (Score:2)
Meh. Some minor technical modifications, and it could have been done.
Attach a trailer with fuel in it, and run a line to the gas tank.
Re: (Score:2)
Meh. Some minor technical modifications, and it could have been done.
Attach a trailer with fuel in it, and run a line to the gas tank.
Or put it on a Tesla once these [youtube.com] have made it to production.
Re: Color me skeptical (Score:1)
At least you're consistent.
Re: (Score:2)
No, it's actually quite a valid criticism. To quote Lyingdowski:
"If there was nobody in the car, it would have worked," he said.
The only stupid one is you. The guy clearly claims the car could have done the trip with no one in the car which is frankly not possible.
Re: (Score:2)
Re: (Score:2)
Re: (Score:1)
So the car was also going to clean its own cameras as well?
Re: (Score:1)
Awww, some butthurt AC doesn't like me. I'm just so devastated! Not
Re: (Score:2)
I've skimmed a couple of gas stops/sleeps and I can't tell if it can actually navigate to a parking spot or gas station.
It seems to start and end on limited access highways too. If it cannot get on and off a limited access highway I'd think it's being exaggerated in its abilities.
I'd suspect Tesla could do this too given some tries.
I saw nothing like road work, or even stopped traffic.
Though if it's camera only, that's something.
Re: (Score:2)
Yea. And I noticed a few times where it seems to start from the shoulder.
1:20 and 1:37 of note.
I think that the AI handled the "on highway" portion, which is the easy part. Obviously it couldn't handle the "pumping gas" part. There are a few single lane pieces where I'm not sure what was going on since they're not highway.
reckless not Wreckless. And no night driving. (Score:2)
So if it only used video camera's it was not driving at night.
Nor was it being placed in situations were there were indecipherable objects in the roadway.
Video can't judge object size or reliably get distance to an object unless it has some references.
For example, a hefty bad in the middle of the road. Now if the road is straight and the lines are dashed or even spaced so that the system can estimate paralav then it could figure out the extent of the hefty bag partially. It will not be able to separate dist
Re: (Score:2)
What if it had 2 cameras in the same direction?
Re: (Score:2)
it doesn't help much for distant objects.
Re: (Score:2)
What if it had 2 cameras in the same direction?
it doesn't help much for distant objects.
So your objection is that it can't do something that humans can't do, and that somehow renders it unable to drive? With greater stereo separation, you get greater depth perception. Our eyes are just a few inches apart. In a car, you can place the cameras feet apart. An AV's ability to determine depth from cameras alone can be superior to ours, and ours is good enough for driving. QED, this is not a real limitation.
I'm not saying he has done it, mind you. I'm saying it's conceivable.
I can drive with one eye
Re: (Score:2)
Video can't judge object size or reliably get distance to an object unless it has some references.
It should provide at least the same possibility to judge those as a human does with only visual input, which seems to be enough to drive a car from coast to coast.
However, I find it rather unlikely that this guy has *algorithms* that can make those judgements to the same level as a human.
Meaningless (Score:2)
This so-called 'engineer' is a danger to himself and others and having admitted what he's done they should find something to prosecute him for.
How does he not have a "store of trust"? (Score:2)
Just because Uber's car had some software disabled that would have prevented collusion with a pedestrian, how does the mean this guy lacks trust?
Uber's cars were working generally OK - on city streets mind you - until they were pulled, because of one accident.
I actually don't find it very hard to believe this could be done myself, because you are talking about almost all city, or near highway driving which is generally straightforward. Pretty much no obstacles, maybe some construction zones which are usual
Trust in Ability vs. Trust in Ethics. (Score:2, Insightful)
He doesn't lack a store of trust because of the accident, he lacks the store of trust because he's a thief.
So a guy who is highly desired by TWO companies for his autonomous driving knowledge cannot be "trusted" to have managed to drive cross country in one?
I wouldn't trust him very much in terms of a contract, but he obviously has the TECHNICAL skills to do what he claims. And there is video of course.
Like I said, what he did was not even that TECHNICALLY challenging, so it's more likely than not he did w
Re:Trust in Ability vs. Trust in Ethics. (Score:5, Informative)
Having drive across the USA on probably similar roads, I can tell you that this isn't really a test, even if he did do it.
In that 3000 miles, everyone was going the same direction with multiple lanes for most all of the way. There were no pedestrians, no animals, no left turns, no stop lights, no school buses, no varying speeds in lanes for most of the distance, probably good weather and no random variations. Were there interesting obstacles, I'm sure they'd be pointed out in the video for their points-value.
So is this a real trial, or just PR? I say: PR. Nothing to see here, move along, sort of stuff.
Re: (Score:2)
I had the same thought. My car has lane keeping and radar cruise control. That's pretty much all you'd need to make a nice all-highway continental crossing where you take control for "scheduled stops."
Re: (Score:2)
Yes that is what I am saying (Score:2)
I can tell you that this isn't really a test, even if he did do it.
That is exactly what I am saying, it's not really that much of a test of todays autonomous driving cars because pretty much all highway driving is easy.
I do think the car probably drove itself to the gas stations and hotels used - that's a bit more impressive, but also like I said, roads near highways tend to be pretty wide, and clearly marked.
Pure city driving or really bad weather is more where things get interesting I think.
Even though yo
Re: (Score:2)
Real tests? Bad weather, heavy diagonal/lateral traffic, density at speed, high speed differentials, random variables like bad drivers/drunks/over-the-hill/dealing-with-children/texters, low vision halt obstacles (school buses, crossing guards, construction traffic), and more--- with mixes of these.
Add in points for slick pavement (including black ice, snow, fresh rain after droughts bringing oil to the surface, wet leaves, gravel, mixtures of these), darting dogs, pavement irregularities (including flats a
Re: (Score:2)
Add in points for slick pavement (including black ice, snow, fresh rain after droughts bringing oil to the surface, wet leaves, gravel, mixtures of these), darting dogs, pavement irregularities (including flats and anomalous breakdowns), children, food trucks, postal vehicles, delivery trucks (human-controlled and not), and it's an evil brew.
Autonomous cars already handle all of these situations much better than humans, as they have better sensors, reaction times, and understanding of how to handle slippery
Re: (Score:2)
Re: (Score:3)
Hop on I-80. Goes coast to coast. Sure, the limits go up and down. Even my lousy TomTom GPS knows what they are. Put your car on cruise, and you can steer only for hundreds and hundreds of miles. Straight stretches across Iowa, Nebraska, Wyoming, and especially parts of Utah and Nevada..... cruise.
I've done stretches of it with a truck and trailer, diesel & TT. Same answer. Cruise. Faster where possible, some braking where not. Cities, merge lanes, pretty easy. Yeah, some occasional movements, always pa
Re: (Score:2)
Yes, exactly. We have had technology capable of driving a vehicle that far on long, wide roads probably as far back as the middle of the last century.
There are so many edge cases that they make up most of the difficulty in autonomous navigation, IMHO. I bet this new car a) would not recognise a cop flagging it down and/or be able to refuel itself. So not much progress really...
Re: (Score:1)
Are you being intentionally dense or are you just an idiot?
Re: (Score:3)
Just because Uber's car had some software disabled that would have prevented collusion with a pedestrian, how does the mean this guy lacks trust?
Are you saying the pedestrian was in on it?
Re: (Score:2)
The New York Times reports that Uber's autonomous vehicles require human intervention every 13 miles, on average, while Google's go 5,600:
https://boingboing.net/2018/03... [boingboing.net]
About as believable.. (Score:1)
Re: (Score:1)
huh? driving is not the pinnacle of human achievement. This is a very doable task. God, I hope that in 5 years we are laughing about how people used to have to drive cars themselves. There will be a lot more of us laughing too since many of us won't be dead from the car crash epidemic that plagues us currently since humans are pretty terrible drivers.
Re: (Score:2)
Re: (Score:2)
Re: (Score:1)
I use comma.ai and their opensource platform OpenPilot. It works. It's here today. It does not know how to handle crazy events, like a deer running back and forth across the street (it will try to avoid and then it will stop and won't go until the driver takes over).
What is not here is autopilot for the masses. A competent driver is required to be at the controls, that is why a system like OpenPilot will never be a platform. Too many idiots raises the cost of liability. However crowd-sourced insurance
Re: (Score:2)
Yeah, it's basically first-gen tesla autopilot.
it can stay in a lane and react to changes in traffic speed.
Of course it'll work SOMETIMES. (Score:5, Insightful)
Self-driving works the vast majority of the time. How many attempts were made (by him and/or others) that we're not hearing about because they had to be aborted? Just doing it once is not exactly Lewis & Clark territory here.
Re: (Score:2)
2 other times. To be fair being pulled over by the cops for not-speeding (and therefore being the classic cautious-buzzed-driver) wasn't their fault. The other time was high-winds blowing it out of a lane. Which seems like a relatively easy enough fix. Probably exceeded its 'safe-steering' limits.
Re: (Score:2)
Re: (Score:2)
Right, but the price of failure is not "You have died of dysentery". There is some chance the car could crash, but it's less than the risk of a human driver crashing a car on a sleep-deprived long drive. It's not even "crashing this plane with no survivors". It would take deliberate malfeasance to make a self-driving car more dangerous than one piloted by your average human. (I don't care what race drivers can do, because very few of us will ever get there.)
Even the Wright brothers risked death or serious i
Nothing new here (Score:2)
Levandowski told the Guardian that, although he was sitting in the driver's seat the entire time, he did not touch the steering wheels or pedals, aside from planned stops to rest and refuel.
So not any different from driving with cruise control and lane assist, then.
Re: (Score:2)
But if you had cruise control and lane assist that were trustworthy and legal to NOT have to pay attention (read a book, browse the net, sleep, be prepared to take over with a 30 second warning) that would be enormously useful, even if it only worked on highways. I hope we are not too far off that. Obviously cars without steering wheels are another thing entirely.
Re: (Score:2)
But if you had cruise control and lane assist that were trustworthy and legal to NOT have to pay attention (read a book, browse the net, sleep, be prepared to take over with a 30 second warning) that would be enormously useful, even if it only worked on highways. I hope we are not too far off that. Obviously cars without steering wheels are another thing entirely.
Well, this is what the traditional auto-makers are aiming for, and already have under legal testing in Europe. The Silicon Valley giants aim higher, and aims to pull off the cars without the steering wheel.
Highway easier (Score:2)
Re: (Score:2)
No LIDAR? (Score:1)
It didn't use lasers? Sometimes a big "spot" in the road is merely discoloration, spilled paint, or a reflection. Such could easily fool camera-driven AI to slam on the breaks, risking a rear-end collision. Lasers can verify such a spot is "flat" in a more direct manner.
I've had close calls myself over mistaken identity because reflections etc. confused (human) stereo vision, being one eye may catch a reflection that the other eye doesn't, a
Re: (Score:2)
The first two times I assume he had to intervene and take over. I doubt it has the ability to perform an emergency stop.
This was his third attempt.
Look at the video (Score:2)
It's basically first-gen Tesla auto-pilot.
There's no footage of it doing anything but staying in one lane.
Nothing of it pulling in to a gas station, nothing of it navigating an interchange.
It seems to not even be able to change lanes.
Re: (Score:1)
It clearly changes lanes quite a few times in the video. In the upper right corner it even says what it is doing (Driving, Merging, Changing Lanes, Construction Driving).
Agree there is no footage outside of freeway/highway driving. Would have been especially interesting in the SF and NYC sections. It also zips through the middle of the USA ridiculously fast. I guess that was boring flat corn fields.
Not sure if it is fair to compare it to Tesla's first-gen autopilot or not.
Obligatory xkcd quote (Score:2)
I-80 (Score:2)
It would not be difficult to build an autonomous vehicle that could drive coast-to-coast on I-80. The autonomous systems company I worked for in the oughts could have built such a vehicle in the 1970s. Except of course for those tricky bits at the endpoints, San Francisco and New York City, I'd suggest it not trying to go through the Chicago metro area at rush hour either even though I-80 is fairly far south. So yeah, possible, easy even - except for the hard part.
Re: (Score:2)
Re: (Score:2)
Whoosh!
Coast to coast seems easier than inner city (Score:2)
Coast to coast is all highway driving. Still impressive, but I would think that getting around in a congested city would be more challenging.
On the highway you have no bicyclists, or dogs, jumping in front of you. No traffic lights, no stop signs. Nobody running stop signs or traffic lights.
He is either lying (Score:3)
Wooo.... (Score:1)