Tag Archives: driverless cars

We don’t want automated cars driving the current speed limits

Posted on by

Should automated vehicles follow all the current traffic laws or will they need to be changed?

When Delphi took its prototype Audi robocar from San Francisco to New York in April, the car obeyed every traffic law, hewing to the speed limit even if that meant impeding the flow of traffic.

“You can imagine the reaction of the drivers around us,” Michael Pozsar, director of electronic controls at Delphi, said at a conference in Michigan last week, according to Automotive News.  “Oh, boy. It’s a good thing engineers have thick skin. All kinds of indecent hand gestures were made to our drivers.”

And that indicates that a problem is brewing, argues Prof Alain Kornhauser, who directs the transportation program at Princeton University. “The shame of the driving laws is that they all sort of have a ‘wink’ associated with them,” he says. “It says 55 miles per hour, but everyone knows that you can do 9 over. If that’s the situation, why isn’t it written that way—with a speed limit at 64?”…

In fact, if all cars were autonomous and connected to each other wirelessly, they wouldn’t need stop signs even at the intersections of multilane highways…

I imagine following the speed limit in the Chicago area would lead to some very unhappy drivers. Theoretically, we might not even need speed limits with driverless vehicles as it would all be dependent on the conditions. This might mean that vehicles would go slower at times than drivers might like (perhaps in inclement conditions) but could go a lot faster and more safely even with a good number of drivers nearby.

But, if traffic laws need to be changed, when exactly would this happen? Presumably, it will take some time to introduce these vehicles onto the road and some time for them to grab a large part of the market. Of course, the government could push all new cars in this direction – particularly since they could be so much safer – but older models would still be on the road for some time. To change the laws, all the cars need to switch over at once, an unlikely event. Until all cars are driverless, traffic laws would have to be more conservative to account for drivers but that probably wouldn’t make the new owners happy.

Overall, I haven’t seen much discussion of how automated cars and cars with drivers will mix even as we creep closer and closer to this eventuality.

Considering a robot superhighway from Mexico to Canada

Posted on by

If driverless cars are in the near future, why not a superhighway of autonomous trucks linking Mexico and Canada?

The project is currently being considered by members of the Central North American Trade Corridor Association (CNATCA), and would consist of a robot-only corridor running along Route 83 through Texas, Oklahoma, Kansas, Nebraska, South Dakota, North Dakota and on into Manitoba.

One of the main reasons for a robot road like this, according to Marlo Anderson of the CNATCA, is that North Dakota produces a lot of oil right now, and doesn’t have a great way to get it all where it needs to go. Sure, there are trains, but there’s not enough space to be had. That, and the jury-rigged cars that carry the oil keep exploding. Trucks can help ease the pressure, especially if they don’t need drivers…

There are plenty of problems to solve before any of this would be possible though, including self-driving car laws in half a dozen US states, some way of having driver-less robo-rigs cross borders into and out of the United States, and security in place to make sure no one tries to exploit that system. But robot roads like this one—if it happens—could pave the way to wider acceptance of self-driving vehicles that really do take care of it all themselves. Even if we’re not ready to have them on the road with us just yet.

Advantages include safer roads, no time restrictions on the trucks, lower labor costs, and presumably cheaper goods and/or more money to be made. Disadvantages include lost trucking jobs, a long period of time to put this all together, and perhaps the biggest hurdle for now: what exactly would such a highway cost to build and maintain? Do we need a fleet of herding vehicles to service the trucks and highway?

I wonder what the final arguments regarding this might look like: perhaps safety on the trucking side (how can you argue with a safer driving experience?) versus the steady erosion of jobs greased by free trade (this time to autonomous vehicles).

Expedia’s 2015 Road Rage Report

Posted on by

It may be all about marketing but a new survey from Expedia looks at what makes American drivers mad:

Data from this year’s effort indicated that “The Texter” generates the most fury among pre-established categories of American drivers, earning scorn of 26 percent of 1,000 respondents.

Rounding out the top five, “The Tailgater” (13 percent) ranked second, narrowly edging out “The Left Lane Hog” (12 percent), “The Crawler” (10 percent), and one of my least favorites “The Multitasker (7 percent)…

In addition to evaluating the most deplorable driver behaviors, the report also found that the least popular in-car behavior is “back-seat driving,” cited as the biggest pet peeve by 52 percent of survey respondents.

The “Reluctant Co-Pilot”—the passenger who won’t help navigate—ranked second (12 percent), followed by the “Radio Hog” (10 percent), “The Snoozer” (8 percent), and “The Shoe Remover” (7 percent)…

Respondents offered multiple reasons for driving misbehaviors including running late and being provoked by other drivers. Rudeness behind the wheel also can be attributed in part to where people drive—the Expedia 2015 Road Rage Report indicated that New York City (42 percent) and Los Angeles (32 percent) topping the list.

This is why we need self-driving cars: these kinds of road rage emotions will disappear. Of course, other kinds of road rage could develop such as getting angry or annoyed with your autonomous vehicle (akin to getting mad at your web browser that takes two seconds to load a page), still getting mad at traffic and congestion (couldn’t driverless cars actually increase traffic?), and the boredom that comes with sitting in a vehicle without having to drive (though you could do more work).

Self-driving semis to bring safety, limit unwanted jobs – and lower the costs of products?

Posted on by

Wired sums up some of the advantages autonomous semis might offer but leaves off a third possible advantages: cheaper shipping costs which leads to cheaper goods.

In 2012 in the US, 330,000 large trucks were involved in crashes that killed nearly 4,000 people, most of them in passenger cars. About 90 percent of those were caused by driver error. “Anything that can get commercial vehicles out of trouble has a lot of value,” says Xavier Mosquet, head of Boston Consulting Group’s North America automotive division.

So it’s no surprise some of the country’s largest freight carriers have in recent years started equipping their vehicles with active safety features like lane control and automatic braking. The economic case for these measures—the predecessors to fuller autonomy—is clear, says Noël Perry, an economist who specializes in transportation and logistics…

Another point in favor of giving robots control is the serious and worsening shortage of humans willing to take the wheel. The lack of qualified drivers has created a “capacity crisis,” according to an October 2014 report by the American Transportation Research Institute. The American Trucking Associations predicts the industry could be short 240,000 drivers by 2022. (There are roughly three million full-time drivers in the US.)

That’s partly because long haul trucking is not an especially pleasant job, and because it takes time and money to earn a commercial driver’s license. The shortage will get worse, Perry says, thanks to a suite of regulations set to take effect in the next few years. A national database to collect company-performed drug and alcohol tests will make it harder for drivers who get in trouble at one job to land another. Speed limiters could keep trucks to a pokey 64 mph. Mandated electronic reporting of hours driven will make it harder to skirt rest rules and drive longer than allowed. These are all good changes from a safety perspective, but they’re not great for profits.

Safety is good and more meaningful jobs might be helpful – though losing a bunch of driving jobs won’t look good to many. But, what about the added benefit of cheaper shipping costs in the long run? Perhaps it will take some time for this technology to become cheap and widely adopted. Yet, if trucks can drive themselves and drivers don’t need to be paid, can’t these trucks run all day long making runs back and forth? And imagine if they could utilize greener technologies as well, limiting fuel costs. Americans like their cheap consumer goods and having everything shipped by semi just a little bit cheaper on store shelves may help Americans enjoy self-driving trucks even more.

Drivers, companies adjusting to changes in car insurance due to autonomous vehicles

Posted on by

A recent survey asked Americans why they would buy an autonomous car and cheaper car insurance was second on the list:

Of the 1,500 US drivers the Boston Group surveyed in September, 55 percent said they “likely” or “very likely” would buy a semi-autonomous car (one capable of handling some, but not all, highway and urban traffic). What’s more, 44 percent said they would, in 10 years, buy a fully autonomous vehicle…

The leading reason people are considering semi-autonomous vehicles isn’t greater safety, improved fuel efficiency, or increased productivity—the upsides most frequently associated with the technology. Such things were a factor, but the biggest appeal is lower insurance costs. Safety was the leading reason people were interested in a fully autonomous ride, with cheaper insurance costs in second place. (Reasons not to want a robo-ride include fear of hacking, distrust of the technology, and good old love of driving.)

This is unexpected, because how insurance will shake out usually is on the “tricky things to be figured out” side of the ledger, alongside how the government will test and regulate the vehicles. The current insurance business model—car owner has insurance to protect himself from the risk of causing a crash—doesn’t make sense if the computer’s in charge. And if we can make cars that rarely crash, do we even need insurance? We certainly won’t need to spend as much on it (currently about $800 a year, according to the National Association of Insurance Commissioners)…

So yes, we’ll be rewarded financially for giving up the wheel. But in the long run, as fully autonomous cars take over our roads, the insurance companies will have to adapt. They can’t argue against saving lives, but “they’re very, very concerned,” says David Carlisle, chairman of the board of auto industry consultancy Carlisle & Company. “If the car can’t wreck anymore, those premiums have got to go down drastically.”

Sounds like a shake-up is coming for the car insurance industry. And if the business becomes a lot less profitable, how many firms will want to participate?

Another thought for those potential buyers of autonomous cars: how long would it take in car insurance savings to make up for the extra technology needed in the car? This could be like the current hybrid or electric car situation where the premium for such vehicles would take years in gas savings to cancel out.

What will we all do in autonomous cars? Use 4G, wireless connections

Posted on by

With the rapidly approaching autonomous vehicles, how will we spend the time once devoted to driving (or backseat driving)?

According to the U.S. Census Bureau, approximately 105 million Americans commute by car each day. With an average round trip of 50 minutes, that’s a whopping 88 million hours daily. “I want some of these hours back,” said Cooley. At this point, “we’re now already moving into the era of post-self-driving vision. It’s not so much about the technology of self-driving, but what will we do after that.”…

The most important groundwork, or “plumbing,” as Cooley refers to it, is for auto manufacturers to integrate 4G into vehicles. Built-in high-speed wireless connectivity makes a car much like a phone, Cooley said, allowing users to get into the habit of using maps, calls, notifications, and other interactive elements in the car.

Given today’s world, this isn’t too surprising: cars will free up even more time for Internet and social media usage. Indeed, we’ll have extra time for multitasking where we can listen to music or watch something while participating online (while being a passenger). Are there better things we could do? With all the studies on sleep deprivation (partly due to media usage), perhaps not driving should lead to more sleep. Or, employers and workers might do more job-related activities on the way to and from work.

Alas, all of these supposed time-saving devices may just keep providing opportunities to do more work or entertain ourselves further…

23-acre city for autonomous cars and “mechatronic pedestrians”

Posted on by

A new University of Michigan facility is ready to test autonomous cars in urban settings:

Welcome to M City, a soon-to-open 23-acre mini-metropolis at the University of Michigan, where automakers can test autonomous cars to prepare for the driverless future expected within a decade. Seeking to replicate a modern city’s chaos—traffic jams, unpredictable pedestrians, weaving cyclists—M City starts running on July 20 and has carmakers and tech companies queuing up to conduct research on its roads…

M City sits amid towering pines in the Detroit suburb of Ann Arbor, a short hop from the technology labs of multiple carmakers. Once completed this summer, the $6.5 million facility will be outfitted with 40 building facades, angled intersections, a traffic circle, a bridge, a tunnel, gravel roads, and plenty of obstructed views. There’s even a four-lane highway with entrance and exit ramps to test how cars without a driver would merge.

“Mechatronic pedestrians” who occasionally pop out into traffic will provide a critical—and bloodless—measure of whether sensors and automatic brakes can react in time to avoid running down a real person. As in a Hollywood backlot, building facades can be rearranged to add to the chaos confronting the chip-controlled vehicles…

Eventually, hundreds of robot cars will ply M City’s urban byways in all seasons and weather conditions. “We would never do any dangerous or risky tests on the open road, so this will be a good place to test some of the next technology,” says Hideki Hada, general manager for electronic systems at Toyota’s Technical Center in Ann Arbor. “A big challenge is intersections in the city, because there are vehicles, pedestrians, and bicycles together with complex backgrounds with buildings and connections to infrastructure. That’s why this is really important.”

I wonder if the results will get released. Just how random will those “mechatronic pedestrians” act in order to replicate the unpredictable nature of human beings? Will the landscape eventually involve real humans (given the use of undergraduate students in psychology experiments, there is a large potential pool of participants nearby)? How real will the fake city look, particularly if this becomes an area for photo opportunities?

For some reason, it sounds like the sort of facility that could easily become the setting for a horror/sci-fi film where automated cars and people come to life and wreak havoc on the urban landscape…

McKinsey predicts drop in car ownership (after 2040) due to self-driving cars

Posted on by

McKinsey suggests one side effect of self-driving cars will be less need for owning one:

But it’s in Phase 3, after 2040, that the fun begins. This is the point where autonomous cars become our primary means of transport, and all the rules are up for debate. Just as car design will fundamentally change once things like forward-facing seats, mirrors, and pedals are no longer necessary, the way we structure physical space could evolve: McKinsey predicts that by 2050, we might need just 75 percent of the space we now reserve for parking our cars. Because this is America, that means we get back 5.7 billion square meters of space—enough to hold the Grand Canyon and then some. That’s because autonomous cars can pack themselves together tightly (no need to allow space for human to exit).

More than that though, our entire idea of car ownership could change. Currently, cars sit unused about 95 percent of the time. That leaves a lot of room for improvement in terms of how we allocate resources.

We won’t stop buying cars altogether—people will still want the option to “independently drive and use the vehicle, and have fun doing so,” says Kaas— but we will buy fewer cars. Without the need for a human at the helm, one autonomous vehicle could take the place of two conventional vehicles: If Joan is going golfing and Joe needs to go shopping, a single car could drop Joan off at the club, swing back to the house to take Joe to the supermarket and back, then return to the club and get Joan. Kaas also predicts you could see the rise of private commuting services, shuttling customers around for a fee.

The recurring theme in the McKinsey report is that the consumer wins. Yes, cars crammed full of high-end technology will likely cost several thousand dollars more than they do today. But “drivers” will save money in the form of regained time (spend your commute working instead of driving!) and many fewer accidents: McKinsey pegs the savings on repair and health care bills alone at $180 billion in the US, predicting a 90 percent drop in crashes.

Cars are expensive so this could theoretically save money (as long as the new autonomous cars have reasonable price tags) and offer more convenience. Yet, it could take a lot to overcome the American love of cars. They aren’t simply about convenience or getting from Point A to Point B (and Americans would always choose mass transit if it were more convenient and effective). It is about other ideas in the American Dream, about freedom and independence and having a status symbol and being mobile. Perhaps by 2040, these things won’t matter as much as we all adjust to autonomous vehicles (and perhaps legislation that makes them the norm for safety’s sake). But, this isn’t just a technological change; this requires some big cultural changes as well.

American cities to have driverless cars by next year

Posted on by

It won’t be long before some major American cities feature self-driving cars:

Automated vehicle pilot projects will roll out in the U.K. and in six to 10 U.S. cities this year, with the first unveiling projected to be in Tampa Bay, Florida as soon as late spring. The following year, trial programs will launch in 12 to 20 more U.S. locations, which means driverless cars will be on roads in up to 30 U.S. cities by the end of 2016. The trials will be run by Comet LLC, a consulting firm focused on automated vehicle commercialization…

He explained that they’re focusing on semi-controlled areas and that the driverless vehicles will serve a number of different purposes—both public and private. The vehicles themselves—which are all developed by Veeo Systems—will even vary from two-seaters to full-size buses that can transport 70 people. At some locations, the vehicles will drive on their own paths, occasionally crossing vehicle and pedestrian traffic, while at others, the vehicles will be completely integrated with existing cars…

In addition to the first test site set for Tampa Bay, they’re looking to implement two more projects in Florida. The Comet team is also planning trials in Greenville, South Carolina and Seattle, Washington, where the 70-person buses will be used in public transit.

At 25 to 40 percent cheaper, the cost to ride the driverless public transit vehicles will be significantly less expensive than traditional buses and trains, according to Mr. Clothier. They’ll also be far less expensive to operate. The vehicles are electric, rechargeable and could cost as low as $1 to $3 to run per day.

I used to think people would find such vehicles disturbing at first but I wonder if people will even take a second long by the time these come along. And, even if they are bothered or it takes some time to get used to, people would definitely like the cost savings, safety, and dependability. Now, whether the cost savings would be passed on to mass transit riders is another matter…

California doesn’t know what safety standards to adopt with driverless cars

Posted on by

Who should certify the safety of driverless cars? California is considering this question:

DMV officials say they won’t let the public get self-driving cars until someone can certify that they don’t pose an undue risk. The problem is that the technology remains so new there are no accepted standards to verify its safety. Absent standards, certifying safety would be like grading a test without an answer key.Broadly, the department has three options: It could follow the current U.S. system, in which manufacturers self-certify their vehicles; it could opt for a European system, in which independent companies verify safety; or the state could (implausibly) get into the testing business…

Manufacturers generally would prefer self-certification. That may be where California ends up, but for now the DMV is exploring independent certification — something that doesn’t exist for driverless cars.

In July, the DMV asked third-party testers whether they’d be interested in getting into the game. The department doesn’t have the expertise to create a safety standard and testing framework, so “the department wanted to get a very good sense of what is out there in the market,” according to Russia Chavis, a deputy secretary at the California State Transportation Agency, which oversees the DMV and requested a deeper exploration of third-party alternatives to self-certification.

 

I can’t imagine California or another US state allowing corporations to do this on their own. Perhaps it would be allowed if they agreed to provide generous payouts if their products failed? Yet, given the hubbub about Toyota and its stuck pedals as well as the Takata air bag scares, this is a public safety issue.

I wonder what the public would want. Americans like progress and like cars. But, there would be some fear regarding the safety of driverless cars until they have some sort of independent certification. And how would Google’s reputation these days affect perceptions of these cars?