Chris Urmson

You are currently browsing articles tagged Chris Urmson.

Of the many consequences, intended and not, of driverless cars, the fate of Uber is one of the least important unless you happen to be an investor in the rideshare company or an interested party hoping to see Travis Kalanick’s oft-irresponsible outfit get driven from the road. More vital will be the disappearance of millions of jobs, the saving of that many lives over time, the impact on the environment, etc.

Still, it’s a fascinating business story. Uber’s massive disruption of the taxi industry may be soon viewed as a staggering, though short-lived, victory, much the way CDs wrecked the market for LPs and cassettes, before being quickly usurped by a better technology.  

When autonomous cars do become a going concern, eventually there won’t be any need for a middle man, and, perhaps, ownership of any kind. The fleets will drive themselves in all senses.

Two new excerpts on the topic are followed by a piece from a retro 1969 National Geographic feature.

From a Forbes article by Chunka Mui about Google driverless guru Chris Urmson’s predictions for the sector:

To the inevitable question of “when,” Urmson is very optimistic. He predicts that self-driving car services will be available in certain communities within the next five years.

You won’t get them everywhere. You certainly not going to get them in incredibly challenging weather or incredibly challenging cultural regions. But, you’ll see neighborhoods and communities where you’ll be able to call a car, get in it, and it will take you where you want to go.

(Based on recent Waymo announcements, Phoenix seems a likely candidate.)

Then, over the next 20 years, Urmson believes we’ll see a large portion of the transportation infrastructure move over to automation.

Urmson concluded his presentation by calling it an exciting time for roboticists. “It’s a pretty damn good time to be alive. We’re seeing fundamental transformations to the structure of labor and the structure transportation. To be a part of that and have a chance to be involved in it is exciting.”•

The opening of Christopher Mims’ smart WSJ piece about the potential fall of Uber:

If Uber Technologies Inc. ever collapses, historians may trace its undoing not to its troubles with labor relations, intellectual property, regulatory conflicts or sexual-harassment allegations, but to technological disruption.

This would be the same technological disruption the company itself pledged to use to upend the auto industry and the $2 trillion a year tied to it.

Less than a year ago, Uber Chief Executive Travis Kalanick described self-driving cars as an “existential” threat to his company, saying that his team must get the technology to market before competitors do, or at least at around the same time. Self-driving vehicles would ultimately be much cheaper to operate than ones requiring human drivers—robots work tirelessly and don’t demand raises. The first companies to roll out fleets of automated taxis could quickly drive their human-powered competition into oblivion.

Uber’s philosophy, both internally and in its pitch to consumers, is that it’s a hassle to own a car. The irony is, for the pay-by-the-ride future of transportation to be realized, someone has to own a lot of cars. Chances are, it won’t be Uber.•

Three months before we reached the moon, a moment when machines eclipsed, in a meaningful way, the primacy of humanity, National Geographic published the 1969 feature “The Coming Revolution in Transportation,” penned by Frederic C. Appel and Dean Conger. The article prognosticated some wildly fantastical misses as any such futuristic article would, but it broadly envisioned the next stage of travel as autonomous and, perhaps, electric.

The two excerpts I’ve included below argue that tomorrow’s transportation would in, one fashion or another, remove human hands from the wheel. The second passage particularly relates to the driverless sector of today. Interesting that we’re skipping the top-down step of building “computer-controlled” or “automated” highways, something suggested as necessary in this piece, as an intensive infrastructure overhaul never materialized. We’re attempting instead to rely on visual-recognition systems and an informal swarm of gadgets linked to the cloud to circumvent what was once considered foundational.

· · ·

“People Capsule”: Dial Your Destination

Everywhere I found signs that a revolution in transportation is on the way. 

The automobile you drive today could probably move at 100 miles an hour. But you average closer to 10 as you travel our clogged city streets.

Someday, perhaps in your lifetime, it could be like this….

You ride toward the city at 90 miles an hour, glancing through the morning newspaper while your electrically powered car follows its route on the automated “guideway.”

You leave your car at the city’s edge–a parklike city without streets–and enter on the small plastic “people capsules” waiting nearby. Inside, you dial your destination on a sequence of numbered buttons. Then you settle back to reading your paper. 

Smoothly, silently, your capsule accelerates to 80 miles an hour. Guided by a distant master computer, it slips down into the network of tunnels under the city–or into tubes suspended above it–and takes precisely the fastest route to your destination.

Far-fetched? Not at all. Every element of that fantastic people-moving system is already within range of our scientists’ skills.

· · ·

Car-trunk Computer Issues Orders

Consider automated cars–and when you do, look at the modern automobile. Think of the rapid increase, in the past decade, of electric servomechanisms on automobiles. Power steering, antiskid power brakes, adjustable seats, automatic door locks, automatic headlight dimmers, electronic speed governors, self-regulated air conditioning.

Detroit designers, already preparing for the day your vehicle will drive itself, are getting practical experience with the automatic devices on today’s cars. When more electric devices are added and the first computer-controlled highways are built, the era of the automated car will be here.

At the General Motors Technical Center near Detroit, I drove a remarkable vehicle. It was the Unicontrol Car, one step along the way to the automated family sedan.

In the car a small knob next to the seat (some models have dual knobs) replaced steering wheel, gearshift lever, accelerator and brake pedal.

Moving that knob, I learned, sends electronic impulses back to a sort of “baby computer” in the car’s trunk. The computer translates those signals into action by activating the proper servomechanism–steering motor, power brakes, or accelerator.

Highways May Take Over the Driving

Simple and ingenious, I thought, as I slid into the driver’s seat. Gingerly I pushed the knob forward. Somewhere, unseen little robots released the brake and stepped on the gas.

So far, so good. Now I twitched the knob to the left–and very nearly made a 35-mile-an-hour U-turn!

But after a few minutes of practice, I found that the strange control method really did feel comfortably logical. I ended my half-hour test drive with a smooth stop in front of a Tech Center office building and headed upstairs to call on Dr. Lawrence R. Hafstad, GM’s Vice President in Charge of Research Laboratories. 

The Unicontrol Car–a research vehicle built to test new servomechanisms–is easy to drive. Still, it does have to be driven. I asked Dr. Hafstad about the proposed automated highways that would relieve the driver of all responsibilities except that of choosing a destination.

“Automated highways–engineers call them guideways–are technically feasible today,’ Dr. Hafstad answered. “In fact, General Motors successfully demonstrated an electronically controlled guidance system about ten years ago. A wire was embedded in the road, and two pickup coils were installed at the front of the car to sense its position in relation to that wire. The coils sent electrical signals to the steering system, to keep the vehicle automatically on course.

“More recently, we tested a system that also controlled spacing and detected obstacles. It could slow down an overtaking vehicle–even stop it, until the road was clear!”

Other companies are also experimenting with guideways. In some systems, the car’s power comes from an electronic transmission line built into the road. In others, vehicles would simply be carried on a high-speed conveyor, or perhaps in a container. Computerized guidance systems vary, too. 

“Before the first mile of automated highway is installed,” Dr. Hafstad pointed out, “everyone will have to agree on just which system is to be used.”• 

Tags: , ,

images (16)

In a Backchannel piece, Steven Levy shares everything most things he learned during an inside look at Google’s autonomous-car mission command at the decommissioned Castle Air Force Base in Atwater, California. Most of the (non-)drivers hired to put miles on the vehicles are recent Liberal Arts grads who test the prototypes on streets in Mountain View and Austin. Some are even employed as human props, known as “professional pedestrians.” “We just have to learn to trust,” one tells Levy. It seems the tight-lipped company’s testing of the cars may have gone beyond what people realize.

An excerpt:

Google’s ultimate goal, of course, is to make a transition from testing to systems where no safety drivers are needed — just passengers. For some time, Google has been convinced that the semiautonomous systems that others champion (which include various features like collision prevention, self-parking, and lane control on highways) are actually more dangerous than to the so-called Level Four degree of control, where the car needs no human intervention. (Each of the other levels reflects a degree of driver involvement.) The company is convinced that with cars that almost but don’t drive themselves, humans will be lulled into devoting attention elsewhere and unable to take quick control in an emergency. (Google came to that conclusion when it allowed some employees to commute with the cars, using autodrive only on premapped freeways. One Googler, perhaps forgetting that the company was capturing the whole ride on video, pretty much crawled into the backseat for a phone charger while the car sped along at 65 miles per hour.)

Google also believes that cars should be able to move around even with no humans in them, and it has been hoping for an official go-ahead to begin a shuttle service between the dozens of buildings it occupies in Mountain View, where slow-moving, no-steering-wheel prototypes would putter along by themselves to pick up Googlers. It was bitterly disappointed when the California DMV ruled it was not yet time for driverless cars to travel the streets, even in those limited conditions. The DMV didn’t even propose a set of requirements that Google could satisfy to make this happen. Meanwhile, Elon Musk, CEO of Tesla, is barreling ahead, introducing a driverless feature in his Tesla cars called Summon. He predicts that by 2018, Tesla owners will be able to summon their cars from the opposite coast, though it’s a mystery how the cars would recharge themselves every 200 or so miles.

But maybe Musk is not the first. When I discussed this with [program director Chris] Urmson, he postulated that in most states — California not among them — it was not illegal to operate driverless cars on public streets. I asked him whether Google had sent out cars with no one in them to pick up people in Austin. He would not answer.•

Tags: ,

One of the really wonderful online publications to pop up recently is Steven Levy’s Backchannel, which is full of interesting ideas about our technological world, how we got here and where we’re headed.

Case in point: After 1.7 million miles of road tests, Chris Urmson, director of Google’s self-driving program, has written a piece about the search giant’s foray into completely remaking transportation and reorganizing cities, reducing traffic and pollution. 

Of course, it’s not Urmson’s job to worry about the societal upheaval that will arise should autonomous vehicles be perfected. That will be up to you and I. While robocars will likely save lives, they will kill so many jobs. If new industries don’t emerge to replace these disappeared positions, how do we proceed? It’s not about holding back progress but dealing with disruption in an intelligent and equitable manner. 

Anyhow, Urmson reports fewer than a dozen fender benders thus far for Google’s driverless cars, with all being caused by human drivers. His analysis unsurprisingly favors a driverless future, but it would be pretty widely reported if he was characterizing the safety record inaccurately. An excerpt:

If you spend enough time on the road, accidents will happen whether you’re in a car or a self-driving car. Over the 6 years since we started the project, we’ve been involved in 11 minor accidents (light damage, no injuries) during those 1.7 million miles of autonomous and manual driving with our safety drivers behind the wheel, and not once was the self-driving car the cause of the accident.

Rear-end crashes are the most frequent accidents in America, and often there’s little the driver in front can do to avoid getting hit; we’ve been hit from behind seven times, mainly at traffic lights but also on the freeway. We’ve also been side-swiped a couple of times and hit by a car rolling through a stop sign. And as you might expect, we see more accidents per mile driven on city streets than on freeways; we were hit 8 times in many fewer miles of city driving. All the crazy experiences we’ve had on the road have been really valuable for our project. We have a detailed review process and try to learn something from each incident, even if it hasn’t been our fault.

Not only are we developing a good understanding of minor accident rates on suburban streets, we’ve also identified patterns of driver behavior (lane-drifting, red-light running) that are leading indicators of significant collisions. Those behaviors don’t ever show up in official statistics, but they create dangerous situations for everyone around them.

Lots of people aren’t paying attention to the road.

Tags: ,

Excerpts follow from two posts (one from Andrew McAfee at the Financial Times and one from the TED blog) that look at the progress of driverless cars, which have improved at a stunning pace since theDebacle in the Desert in 2004. Elements of driverless will be helpful, but they change the game in many ways–some wonderful, some concerning–only when they become completely autonomous. McAfee has been further convinced about the sector by recent developments.


From McAfee:

Transportation. Most of us have heard of driverless cars by now. I had the chance to ride in one of Google’s in 2012. It was an experience that went from mind-bending to boring remarkably quickly; the car is such a good and stable driver that I quickly lost all sense of adventure while I was in it. Still, though, I was unprepared for how much progress has been made since then with autonomous road vehicles. Google project director Chris Urmson brought us up to speed with that company’s work, and made a compelling case that we should be striving not for more and better tech to assist human drivers, but instead to replace them. Doing so will save lives, open up opportunity to the blind and disabled and free us from a largely tedious task. And in response to the criticism that self-driving cars aren’t good at dealing with unanticipated events, he showed a video of what happened when one of his fleet encountered a woman in a wheelchair chasing a duck around in the street. The car responded beautifully; we in the audience lost our minds.•

From the TED blog:

Why we need self-driving cars. “In 1885, Carl Benz invented the automobile,” says Chris Urmson, Director of Self-Driving Cars at Google[x]. “A year later, he took it out for a test drive and, true story, promptly crashed it into a wall.” Throughout the history of the car, “We’ve been working around the least reliable part of the car: the driver.” Every year, 1.2 million people are killed on roads around the world. And there are two approaches to using machines to help solve that problem: driver assistance systems, which help make the driver better, and self-driving cars, which take over the art of driving. Urmson firmly believes that self-driving cars are the right approach. With simulations that break a road down to a series of lines, boxes and dots, he shows us how Google’s driverless cars handles all types of situations, from a turning truck to a woman chasing ducks through the street. Every day, these systems go through 3 million miles of simulation testing. “The urgency is so large,” says Urmson. “We’re looking forward to having this technology on the road.”•

Tags: ,

As is their wont, technologists would like driverless cars on the road yesterday, but traditional automakers would rather ease into the sector with assisted-driving functions introduced gradually. A new Economist report is bearish on Silicon Valley’s chances of becoming kings of the road even should the industry go electric and autonomous, citing the nouveau carmakers’ lack of infrastructure (in both manufacturing and corporate) in dealing with many problems inherent to the business. I think the piece’s prediction from Boston Consulting that “cars with even limited self-driving features will never exceed 25% of sales” will only be true if they’re eclipsed by fully autonomous models before surpassing that number. Otherwise most models will probably soon have numerous robocar features at the disposal of human drivers. An excerpt:

The head of Google’s autonomous-car project, Chris Urmson, nevertheless argues that the conventional carmakers’ incremental approach will slow them down, and that a leap straight into fully self-driving vehicles will pay off quicker. However, even if he is proved right in terms of developing the technology, there are two other big barriers to overcome: regulatory approval, and drivers’ nervousness at ceding control entirely to a computer.

Carmakers have had to become adept at handling mountains of regulations and fending off liability lawsuits. These will be huge issues when any self-driving car is involved in an accident—which they will be, even if less frequently than ones driven by humans. Slowly feeding in autonomy may be a better way of convincing road users and legislators of the technology’s benefits. In a pessimistic forecast, the Boston Consulting Group reckons demand for cars with even limited self-driving features will never exceed 25% of sales, and fully autonomous ones will account for just 10% of sales by 2035 (see chart 2).

Perhaps technology firms can accelerate the future of the car. But whatever happens, this is a difficult business to break into. Google would like the carmakers it hopes eventually to supplant to help seal their doom by building its vehicles under contract. Unsurprisingly, none seems too keen on this. Apple’s cash pile of $178 billion is more than enough to set up a carmaking division and tool up its factories. But the technology firms have no manufacturing culture, and the skills needed to market, distribute and provide after-sales service for cars is unlike anything they are used to.•


Reading a new article about Google moving more quickly than anticipated with its driverless dreams reminded me of a passage from a Five Books interview with Robopocalypse author Daniel H. Wilson. An excerpt from each piece follows.


From Five Books:


Isn’t machine learning still at a relatively early stage? 

Daniel H. Wilson:

I disagree. I think machine learning has actually pretty much ripened and matured. Machine learning arguably started in the 1950s, and the term artificial intelligence was coined by John McCarthy in 1956. Back then we didn’t know anything – but scientists were really convinced that they had this thing nipped in the bud, that pretty soon they were going to replace all humans. This was because whenever you are teaching machines to think, the lowest hanging fruit is to give them problems that are very constrained. For example, the rules of a board game. So if you have a certain number of rules and you can have a perfect model of your whole world and you know how everything works within this game, well, yes, a machine is going to kick the crap out of people at chess. 

What those scientists didn’t realise is how complicated and unpredictable and full of noise the real world is. That’s what mathematicians and artificial intelligence researchers have been working on since then. And we’re getting really good at it. In terms of applications, they’re solving things like speech recognition, face recognition, motion recognition, gesture recognition, all of this kind of stuff. So we’re getting there, the field is maturing.

“What those scientists didn’t realise is how complicated and unpredictable and full of noise the real world is. That’s what mathematicians and artificial intelligence researchers have been working on since then. And we’re getting really good at it. In terms of applications, they’re solving things like speech recognition, face recognition, motion recognition, gesture recognition, all of this kind of stuff. So we’re getting there, the field is maturing.•


From Phys. org:

The head of self-driving cars for Google expects real people to be using them on public roads in two to five years.

Chris Urmson says the cars would still be test vehicles, and Google would collect data on how they interact with other vehicles and pedestrians.

Google is working on sensors to detect road signs and other vehicles, and software that analyzes all the data. The small, bulbous cars without steering wheels or pedals are being tested at a Google facility in California.

Urmson wouldn’t give a date for putting driverless cars on roads en masse, saying that the system has to be safe enough to work properly.

He told reporters Wednesday at the Automotive News World Congress in Detroit that Google doesn’t know yet how it will make money on the cars.

Urmson wants to reach the point where his test team no longer has to pilot the cars. “What we really need is to get to the point where we’re learning about how people interact with it, how they are using it, and how can we best bring that to market as a product that people care for,” he said.•

Tags: ,

Google, which knows the real money in the potential driverless-auto sector is in the software, wants to bring 1.0 to market in the next five years with the help of Big Auto. If the timeline the company has laid down is to be realized, the cars will need to be able to “see” to overcome current infrastructure and mapping limitations. From Joseph B. White and Rolfe Winkler at WSJ:

“The biggest challenges for Google’s efforts involve software, not hardware, [Chris] Urmson said. Google is confident, for instance, that it has laser radar technology, or LIDAR, that can provide accurate images of a car’s surroundings at a reasonable cost, he said. …

There are also some important differences between the strategy Google is pursuing to develop and bring its technology to drivers and the way auto makers are approaching automated driving.

Google wants to develop a fully automated car that doesn’t require any input from the driver. Mr. Urmson said it is difficult to get a driver who isn’t paying attention to the road to suddenly—and safely—retake the wheel. Further, he said, a partially automated car ‘doesn’t help a blind man get lunch or help an aging widow get to her social events.’

This is why Google is developing designs that would entail no steering wheel. For now, Mr. Urmson said his team is working on vehicles that would operate at speeds below 25 miles an hour—which would qualify Google’s car as a neighborhood electric vehicle that doesn’t have to be equipped with air bags or meet certain other safety standards required of conventional cars.

One scenario for such vehicles would be to position them in a city’s central business district and allow people to summon the cars with a smartphone app, Morgan Stanley analyst Adam Jonas wrote in a recent report. Mr. Urmson said that is one avenue the company could pursue.”

Tags: , , ,

Autonomous vehicles are probably not as close as we’d like nor as far in the distance as we might believe, but they’re not road-ready yet and hurdling those last few obstacles may be more difficult than all the ones that came before them. From Lee Gomes at Technology Review:

“Would you buy a self-driving car that couldn’t drive itself in 99 percent of the country? Or that knew nearly nothing about parking, couldn’t be taken out in snow or heavy rain, and would drive straight over a gaping pothole?

If your answer is yes, then check out the Google Self-Driving Car, model year 2014.

Of course, Google isn’t yet selling its now-famous robotic vehicle and has said that its technology will be thoroughly tested before it ever does. But the car clearly isn’t ready yet, as evidenced by the list of things it can’t currently do—volunteered by Chris Urmson, director of the Google car team.

Google’s cars have safely driven more than 700,000 miles. As a result, ‘the public seems to think that all of the technology issues are solved,’ says Steven Shladover, a researcher at the University of California, Berkeley’s Institute of Transportation Studies. ‘But that is simply not the case.’

No one knows that better than Urmson. But he says he is optimistic about tackling outstanding challenges and that it’s ‘going to happen more quickly than many people think.'”

Tags: , ,

Google believes its driverless cars are already safer than human drivers. Even if that’s currently too ambitious a statement, it’s really only a matter of time. From Tom Simonite at the Technology Review:

“Data gathered from Google’s self-driving Prius and Lexus cars shows that they are safer and smoother when steering themselves than when a human takes the wheel, according to the leader of Google’s autonomous-car project.

Chris Urmson made those claims today at a robotics conference in Santa Clara, California. He presented results from two studies of data from the hundreds of thousands of miles Google’s vehicles have logged on public roads in California and Nevada.

One of those analyses showed that when a human was behind the wheel, Google’s cars accelerated and braked significantly more sharply than they did when piloting themselves. Another showed that the cars’ software was much better at maintaining a safe distance from the vehicle ahead than the human drivers were.

‘We’re spending less time in near-collision states,’ said Urmson. ‘Our car is driving more smoothly and more safely than our trained professional drivers.'”

Tags: ,

FromLet the Robot Drive: The Autonomous Car of the Future Is Here,Tom Vanderbilt’s new Wired piece about humans on the verge of relinquishing control of the wheel:

“[Chris] Urmson, with the soft-spoken, intense mien of a roboticist who has debugged a Martian rover in the deserts of Chile, occupies the nominal ‘driver’s seat’—just one of the entities open to ontological inquiry this morning.

The last time I was in a self-driving car—Stanford University’s ‘Junior,’ at the 2008 World Congress on Intelligent Transportation Systems—the VW Passat went 25 miles per hour down two closed-off blocks. Its signal achievement seemed to be stopping for a stop sign at an otherwise unoccupied intersection. Now, just a few years later, we are driving close to 70 mph with no human involvement on a busy public highway—a stunning demonstration of just how quickly, and dramatically, the horizon of possibility is expanding. ‘This car can do 75 mph,’ Urmson says. ‘It can track pedestrians and cyclists. It understands traffic lights. It can merge at highway speeds.’ In short, after almost a hundred years in which driving has remained essentially unchanged, it has been completely transformed in just the past half decade.”


Kraftwerk, “Autobahn,” 1975:

Tags: ,