Self-Driving Car Will Make Trip From San Francisco To New York City

An anonymous reader writes with news that Delphi Automotive is undertaking the longest test of a driverless car yet, from the Golden Gate Bridge to midtown Manhattan. "Lots of people decide, at one point or another, to drive across the US. College kids. Beat poets. Truckers. In American folklore, it doesn't get much more romantic than cruising down the highway, learning about life (or, you know, hauling shipping pallets). Now that trip is being taken on by a new kind of driver, one that won't appreciate natural beauty or the (temporary) joy that comes from a gas station chili dog: a robot. On March 22, an autonomous car will set out from the Golden Gate Bridge toward New York for a 3,500-mile drive that, if all goes according to plan, will push robo-cars much closer to reality. Audi's taken its self-driving car from Silicon Valley to Las Vegas, Google's racked up more than 700,000 autonomous miles, and Volvo's preparing to put regular people in its robot-controlled vehicles. But this will be one of the most ambitious tests yet for a technology that promises to change just about everything, and it's being done not by Google or Audi or Nissan, but by a company many people have never heard of: Delphi."

Comments

Or will it?

[istartedi]

Will it, or will it just go off for a few lube jobs and tell us it did that? When AI learns how to cook the books, then I'll be impressed... and scared.

Re:Or will it?

[ShanghaiBill]

AI will not write books, do programming, etc. Strong AI is a myth.

Unless human brains have some magical powers (like a soul blessed by God), there is no logical reason that machines shouldn't eventually be smarter than humans. The only question is how far off it is. There has lately been some big progress on deep neural nets. There is also steady progress in symbolic AI and knowledge ontology. My guess is that we will get there in a couple decades, using a hybrid solution.

AI is just complex automation PROGRAMMED BY HUMANS.

That doesn't mean that it is limited by human intelligence. Good chess programs can easily beat the programmer that wrote them.

Re:Or will it?

[tehcyder]

Unless human brains have some magical powers (like a soul blessed by God), there is no logical reason that machines shouldn't eventually be smarter than humans.

Once you can build a (physical or digital) model that explains why Mozart or Einstein's brains worked as they did compared with the average human being, I'll take comments like this seriously.

My guess is that we will get there in a couple decades

Indeed, like controlled nuclear fusion producing unlimited free energy, True AI is always twenty years off.

Re:If this works, everything will change.

[causality]

Good luck to the Delphi team. Just imagine the possibilities.

The article briefly mentions that there are humans inside [yes, I read it]. Until then, I was wondering how they planned to handle refueling (and maybe in the future, recharging). When they figure that one out, imagine what this kind of system will do to the trucking industry.

Re:If this works, everything will change.

[drinkypoo]

I was wondering how they planned to handle refueling (and maybe in the future, recharging). When they figure that one out, imagine what this kind of system will do to the trucking industry.

They don't even have to figure that one out. All they have to do is institute full service fueling to replace cardlock fuel stops. The trucks pull in, someone wanders over and fuels them, the trucks pull out again when they detect that the tank has been closed and the attendant has moved away. Extensive monitoring and high-resolution cameras will eliminate the need to have a human on board entirely.

Presumably, a car with a telepresence robot in it (just enough to communicate with attendants, not to drive the car) could conceivably already cross the country simply by stopping at full-service stations, if not for the legal impediments.

Besides commuting, OTR trucking really is the "killer app" for self-driving automobiles, and it's coming sooner than people think.

When it works little to nothing will have changed

[burtosis]

The car has a driver. It's nothing more than a slightly longer pre-mapped pre-planned trip with every last detail painstakingly manually entered. The vehicle cannot handle road construction, traffic problems, snow, simple decisions as to what is safer - like running over a blowing plastic bag or slamming on the brakes.
I'm all for progress but this is just one more sensational click bait hype article. It's getting almost as bad as solar cell efficiency articles, which if you took them at face value we would be sitting at around one gazillion percent by now.

Re:Most ambitious

[mrvan]

Highway driving is also the most boring part of driving, and on longer trips often the largest part. City and local driving is kind of fun, you have to pay attention and hopefully you get some nice scenery and usually takes at most 30 minutes. Highways are just boring and can easily take 10+ hours. Fully automated highway driving (even requiring me to stay behind the wheel but letting me read, work or sleep until/unless an alarm goes off) would certainly be a killer feature.

I guess adaptive cruise control plus lane assist or whatever they call it comes close, so it should be possible to get it on the market soonish?

To impress me, try cross-city drives instead.

[wvmarle]

More impressive would be for the car to drive from one end of New York to the other. During the day, avoiding highways, dealing with really chaotic traffic on narrow, poorly marked roads full of distractions and ambiguities.

Highways are simple. Traffic flows in one direction only, clearly marked and wide roads, no intersections, all roughly the same speed. No surprises. It's where by far the fewest accidents happen for human driven cars, even though it's boring and probably the part where human drivers pay least attention. Doing an hour of highways, ten hours of highways, 100 hours of highways - it's just more of the same. Now it's cross country, tomorrow it'll be cross country and back. And back again. As long as the fuel will last.

Re:To impress me, try cross-city drives instead.

[burtosis]

Mod parent up. It's simply another slightly longer preplanned pre-mapped mountain of manually entered data test. The technology cant even do as well as a below average human in perfect sunny day few cars on the road freeway traffic. In a city it would rack up fatalities like a mountain dewed up teenager playing gtaV.

Re:If this works, everything will change.

[jpapon]

This is consistent with the overall American trend of replacing solid blue-collar jobs with entry-level service type jobs.

I fail to see how trucker is a "solid blue-collar job" while "gas station refueler" is somehow an "entry-level service type job". They're both pretty typical blue collar jobs.

I would say adding a full-service attendant at every truck stop gas station is probably the least complicated and easiest-to-implement part of an automated nation-wide self-driving truck shipping system. You're really focusing on the simplest part of the problem.

Re:If this works, everything will change.

[burtosis]

Extensive monitoring and high-resolution cameras will eliminate the need to have a human on board entirely.

Besides commuting, OTR trucking really is the "killer app" for self-driving automobiles, and it's coming sooner than people think.

HaHaHa. Yes you can watch the semi slam on the brakes for a piece of cardboard that gets blown by the wind and creates a multi car pileup. I'm not sure how that helps.
I take it you have never worked on any computer vision papers or any teams that have tried vehicle automation? Because even autonomous freeway driving in the USA is far far far off - perhaps 30 years or more. When you see someone with algorithms that can actually do better than the bottom 5-10% of human drivers (under the same apples to apples situation) then we would have something. That something does not exist today.

Car is not self driving.

[burtosis]

Let's call it for what it is - cruise control plus lane following plus automatic braking to prevent running into vehicles in front of them. It does not turn, it does not make smart safety decisions, it cannot handle simple variations in real conditions. It's as close to being an autonomous car as Siri is to being strong AI.

no it won't.

[YesIAmAScript]

Delphi already said the car will only self-drive portions of the trip. Long portions of the trip, but only some portions nonetheless.

"When it’s not on the highway, one of the humans inside will take the wheel."

Squeegee Guy

[Capt.Albatross]

Just for a moment, I wished the the squeegee guys were back. I would like to see an interaction between a squeegee guy and this car at the exit of the Lincoln tunnel.

Re: CMU: Been There, Done That

[discontinuity]

I just was going to mention the CMU project and someone beat me to it. They drove nearly autonomously from Pittsburgh to San Diego in 1995. Here are some (old) relevant links:

• Project webpage
• No Hands across America Press Release
• Further info with actual dates
• Summary of a 1995 paper about the project

The CMU project was not anything that was near consumer-ready and it also was not 100% autonomous. IIRC, humans had to intervene in more complicated driving scenarios and the autonomous system handled the open highway stuff. they report a figure of 96% or so autonomy by miles driven on a shorter trip from Pittsburgh to DC. So it's not like they had everything figured out back in 1995. But still...1995!

Re:If this works, everything will change.

[drinkypoo]

HaHaHa. Yes you can watch the semi slam on the brakes for a piece of cardboard that gets blown by the wind and creates a multi car pileup. I'm not sure how that helps.

Existing automotive driving assistance systems are already immune to this problem. They use radar to tell the difference between a plastic bag or a piece of cardboard, and something with mass.

Granted, a mylar balloon will probably produce the kind of results you're looking for, but there don't seem to be near as many of those on the highways as pieces of cardboard, so I can see why you didn't go for that example.

Re:If this works, everything will change.

[tompaulco]

This is consistent with the overall American trend of replacing solid blue-collar jobs with entry-level service type jobs. I wonder how long that hypothetical arrangement could last?

Well eventually, as more and more blue collar jobs are replaced with entry level service jobs, nobody will be able to afford the goods and services produced, so the need for OTR trucking will go away, and the companies that make the products will go away, so the need for IT people like us will go away. Everybody wins!

Re:If this works, everything will change.

[Dutch Gun]

I take it you have never worked on any computer vision papers or any teams that have tried vehicle automation? Because even autonomous freeway driving in the USA is far far far off - perhaps 30 years or more.

Cars with autonomous freeway driving will be out in just a couple of years, according to automotive manufacturers. Nearly all the major players are predicting fully autonomous cars will be a solved problem sometime between 2020 and 2025.

Keep in mind that cars can and will make use of a variety of sensors besides vision that are much easier and reliable to process. I'm betting the first generation of automated highway driving uses no vision systems at all... just radar, lidar, and sonar, plus GPS for nativation. More to the point, they can use ALL of them at once. Those are more than enough to handle highway driving. Cars today are already using some of these systems for their intelligent cruise control, auto-parking, and collision avoidance systems.

Make no mistake... computers are going to be FAR better drivers than humans. No getting drowsy or falling asleep. No distractions by the passengers in the seat next to you or the rugrats in the back. No rear-ending cars while looking at your cellphone or putting on makeup or one of the ten thousand stupid things humans do every day behind the wheel. Fully autonomous cars can't come fast enough, and they'll likely be coming a hell of a lot faster than you think.

Re:If this works, everything will change.

[burtosis]

Cars with autonomous freeway driving will be out in just a couple of years, according to automotive manufacturers. Nearly all the major players are predicting fully autonomous cars will be a solved problem sometime between 2020 and 2025.

Keep in mind that cars can and will make use of a variety of sensors besides vision that are much easier and reliable to process. I'm betting the first generation of automated highway driving uses no vision systems at all... just radar, lidar, and sonar, plus GPS for nativation. More to the point, they can use ALL of them at once. Those are more than enough to handle highway driving. Cars today are already using some of these systems for their intelligent cruise control, auto-parking, and collision avoidance systems.

Make no mistake... computers are going to be FAR better drivers than humans. No getting drowsy or falling asleep. No distractions by the passengers in the seat next to you or the rugrats in the back. No rear-ending cars while looking at your cellphone or putting on makeup or one of the ten thousand stupid things humans do every day behind the wheel. Fully autonomous cars can't come fast enough, and they'll likely be coming a hell of a lot faster than you think.

Funny because sensor fusion isn't so simple. It's a big step from auto following a car to actual autonomous driving on freeways in a more reliable way than even the worst human drivers. Comparing the state of the art today to sunny day freely moving traffic human drivers - we are no where near humans. Add heavy rain, dirty sensors, snow, road construction, etc and we will have need for humans to sit and nanny autonomous vehicles from the inside for a long time.
Furthermore all this "solved problem in 10 years" smacks of the leading computer scientists in the 60s claiming they could beat the best human in 10 years. Then it was another 10. Granted it was only 4 times of saying 10 years so yes eventually. Further still there is the legal ramifications of when the first human fatality due to autonomous driving occurs - common sense and logic will take a back seat to "think of the children" in the USA and we both know it. We are more likely 30 years out than 5-10. However what chaps my hide is people saying this stunt of using a combination of cruise control, lane following, and car distance following is autonomous driving is like saying Siri is strong AI.

Cops

[John Jorsett]

Something I've lately been wondering: has anyone yet figured out how to get a robot vehicle to recognize when a cop is behind them with flashing lights and to pull over? Or, if an emergency vehicle is approaching with siren blaring, do likewise? Seems like that would be a good way to hijack the load of a robot truck.

Re:When it works little to nothing will have chang

[romanval]

Actually, the smarter self driving cars CAN handle various obstacles such as road construction and lane closures. See https://www.youtube.com/watch?...

Re:When it works little to nothing will have chang

[burtosis]

If you are talking about the google car video 2:30 into the video - yes cars can just barely - and unsafely - navigate through construction at like 5mph. Google is NOT claiming they can reliably handle construction. Please point me to a statement otherwise. Furthermore in that same video the car is meandering like a texting schoolgirl in its lane - dosen't inspire confidence in me at all. Get that laser rangefinder on top of the car dirty, such as dust from rain and road spray, then tell me how it works. (Answer is the vehicle wont work in rain).

Re:If this works, everything will change.

[Kjella]

The article briefly mentions that there are humans inside [yes, I read it]. Until then, I was wondering how they planned to handle refueling (and maybe in the future, recharging). When they figure that one out, imagine what this kind of system will do to the trucking industry.

When they figure what out? All you need is $5 of electronics to open the fuel tank and a robot arm to hit a stationary, active target that can light itself up like a Christmas tree and say here I am. A month sounds like a long time to solve this problem, more the kind you have a prototype made over the weekend. A year max to create an integrated solution of how to find gas stations & gas prices, locate a pump, two way authentication so nobody can pour sugar in the tank, payment and so on. The only reason we don't solve it today is because it's not a problem worth solving. By the time we have actual self-driving cars that can drive with no passengers - which is still many years away - I'm sure we'll have a few self-tanking stations. And if you want the business from robo-cars, the market will take care of the rest.

Re:If this works, everything will change.

[Dutch Gun]

Have you seen videos of Google's car deftly navigating it's way through city streets, pedestrians, bicycles (even recognizing hand signals), railroads, construction zones, and other such obstacles? They're not claiming it's perfect yet, but I think you're underestimating how close these companies are to a real, working solution.

As far as legal issues... yeah, there will be lawsuits, but that's nothing new for the automotive industry. Self-driving is such a massively compelling feature, though, that it's absolutely going to happen.

I guess we'll just have to wait about a decade to see who's right.

On the road again

[paiute]

Will this car pick me up if I put my thumbdrive out?