Cars That Think iconCars That Think

Germany's Autobahns Say "Willkommen" To Robocars

It must be unnerving for a new driver to prepare to enter the passing lane only to see, at the last possible moment, a metallic monster streaking by. How much more nerve-wracking would it be if the monster had no driver at the wheel? 

For some, that would be the true test of robocar reliability. And, if Germany’s government has its way, it’ll happen first on that country’s autobahn system, famous for its speed-limit-free stretches.

In an announcement earlier this week, the government said that it would soon allow robocars onto autobahn A9, which runs from Berlin straight down to Munich. In the beginning, at least, it will involve experimental cars, such as those now in development at Mercedes-Benz, Audi and BMW

Not all of A9 allows unlimited speed, and it’s unclear whether the early test cars will be given free rein on the parts that do. But the pavement-burning prospect is obviously part of the appeal.

Not only would the cars drive themselves, they’d communicate with other cars and with roadside infrastructure. That means cars that think would be running on roads that think—a technological double-header. And if the German government mandates the use of alternative fuels, it would win the trifecta. Indeed, such a mandate makes sense, given that conserving fuel is one of the stated motives of the new plan.

Another motive is fear of being left behind by, say, Google, and its famous self-driving car. In an interview with the Frankfurter Allgemeine, Alexander Dobrindtsaid, Germany’s Transport Minister, said: “We must guard against the monopolization of the data. We must create for ourselves a digital sovereignty, regardless of the Americas and Asia.”

At least, that’s the English version given by Google Translate. Hmmm, maybe the minister has reason to be on his guard against Google. What’s clear is that the search-engine behemoth has pushed not only companies but counties, states, and countries into action.

Nevada was the first U.S. state to allow Google’s car on its roads, followed by California and other biggies. The Netherlands—famous for its concentration on infrastructure, notably its sea-restraining dikes—is now doubling down on smart cars and roads; last week it announced a plan to open its roads to large-scale testing of autonomous cars and trucks. 

It may at first seem that barreling down the autobahn is the most futuristic aspect of self-driving cars, but in fact, it’s the relatively easy problem, and it will likely be solved quite soon. Far harder is the last-mile problem that robocars face when nosing their way through pedestrian-packed crosswalks and child-infested surburban developments.

Big Brother Is Watching Your Car—in Real Time

Government narcotics agents have secretly built a database of “hundreds of millions of records” on the movement of cars in the United States, the Wall Street Journal reports. The information, originally meant to help steer police toward suspects in drug-smuggling cases, is now used in kidnapping cases and other crimes.

An important point of the program is to confiscate cars and other assets of people suspected of crimes. Such civil forfeiture, as it is called, is controversial in part because it has far fewer safeguards against abuse than punishments made in criminal cases. Whether the car-tracking intelligence program is supervised by the courts remains unclear, the Journal says.

Read More

U.S. Gov’t Wants Automakers to Put Self-braking Cars on the Road

Late last week, U.S. Transportation Secretary Anthony Foxx announced two new additions to the federal department’s New Car Assessment Program (NCAP), which recommends advanced safety features. Foxx said that the National Highway Traffic Safety Administration (NHTSA) will push automakers to include two automatic emergency braking systems that it predicts will save thousands of lives each year.

One, crash imminent braking, applies the brakes if the vehicle senses that a collision is about to happen and the driver fails to act. The other, dynamic brake support, kicks in if the driver does depress the brake pedal, but misjudges how much force to apply. It will supplement the driver's braking effort in order to slow the car before a collision or to at least lessen the severity of the crash.

Read More

Watch this Corvette Z06 Test Us, As We Test It

One part video game, one part reality TV, one part Formula One: The Corvette Z06’s onboard Performance Data Recorder (PDR) not only makes driving better, it also makes for better drivers.

A world’s first on a production car, the PDR records high-definition video and audio of track laps or a scenic weekend cruise. But for the 485-kilowatt (650 horsepower), 315 kph (195 mph) Z06, the PDR is about more than capturing moments. It tracks all sorts of data to power sophisticated track-analysis tools that help drivers improve their performance. 

“If you’re a professional driver, you have tools like this,” says Harlan Charles, the Chevrolet product manager who first envisioned the PDR. “It’s not just about stronger engines and stickier tires, but making yourself better as well.”

I’ll get to what it does and how it works, but first, let me tell you how it feels.

Read More

Has Robocar Safety Been Hyped?

A study out today throws cold water on the accident-free paradise that proponents of autonomous cars have prophecied. 

Not only may robocars never be as safe as the best drivers, their growing pains may temporarily lower overall road safety, say Michael Sivak and Brandon Schoettle, psychologists at the University of Michigan’s Transportation Research Institute. 

They note that some accidents will still be caused when human beings make mistakes for which robotic cars cannot compensate. If, for example, a child darts in front of a car from out of nowhere, even instantaneous braking may not stop the car in time.

Nor is it a foregone conclusion that the strengths of autonomous cars will outweigh those of good drivers.  True, autonomous cars never tire or get distracted, and they can respond to new information almost immediately, but these factors wouldn’t necessarily  “trump the predictive experience of middle-aged drivers.”

Middle age is the sweet spot for safety, because by that stage in life drivers are less likely to take risks and knowledgeable enough to intuit what other people are likely to do. 

Then there is the problem of system failures, which, even in today’s human-piloted cars, cause about 1 percent of fatal accidents. The researchers say that some failures—like a burned-out light—might not bother a self-driving car, which has many overlapping sensor systems. But other kinds of failures may become more common, because the sensing and information-processing hardware is so complex.

Of course, the researchers are trying to correct what they regard as excessive technological optimism. Still, is it entirely fair of them to compare robocars only to the best drivers? Most accidents are caused by the worst ones, and it is beginning to become clear that those are the people that a robot could outperform with one clanky arm tied behind its back.

Google's Self-Driving Car Pals Revealed

Google took the unusual step last week of identifying key technology partners in the development of its new prototype self-driving car. Speaking at the North American International Auto Show in Detroit, project director Chris Urmson named Continental, Roush, Bosch, ZFLS, RCO, FRIMO, Prefix, and LG as companies that had helped to build the pod-like vehicle. The fully autonomous vehicle is undergoing testing at Google’s private track in California prior to road tests at NASA Ames later this year.

Read More

Self-driving Cars Get Good Navigation on the Cheap

Researchers are making rapid progress in developing systems designed to give cars ever more autonomy. The benefits are myriad; and some, like making a dramatic reduction in the tens of thousands of traffic fatalities that occur each year, are priceless. Still, automakers are loathe to add new, costly components without a clear sense of how the additions will justify a higher sales price. Researchers, keenly aware that every penny counts, are looking for ways to cut the cost of making cars smarter.

One example is a new software system from a team at the University of Michigan in Ann Arbor that allows a car to see its surroundings and determine its location using a single video camera instead of several laser scanners.

Lidar, the three-dimensional laser scanning technology used to create real-time maps of a car’s environment that are compared with pre-drawn maps, does the job quite effectively. But Ryan Wolcott, a U-M doctoral candidate in computer science and engineering, has come up with an alternative that he says pinpoints a car’s location with the same level of accuracy as laser ranging but is many times cheaper. Wolcott described his thinking in a paper titled, "Visual Localization within LIDAR Maps for Automated Urban Driving," that was named best student paper at the Conference on Intelligent Robots and Systems in Chicago in September. Wolcott noted that:

"The laser scanners used by most self-driving cars in development today cost tens of thousands of dollars, and I thought there must be a cheaper sensor that could do the same job…Cameras only cost a few dollars each and they're already in a lot of cars. So they were an obvious choice."

Just like navigation systems including the one Google is fine-tuning for its vehicle, the system dreamed up by Wolcott and his collaborator, Ryan Eustice, a U-M associate professor of naval architecture and marine engineering, stores thousands of maps of a given area. But instead of comparing the real-time data to a flood of two-dimensional maps, their system turns the prerecorded data into a series of three-dimensional pictures the way a video game would. This, say Wolcott and Eustice, lets the navigation system compare these 3-D images with the images captured by a conventional video camera as the car cruises along city streets.

They immediately banged up against a problem: how to process the staggering amounts of video data the system would generate, all in real time. Video gaming again inspired the solution. They built in graphics processors that are common to video game consoles.

“One of the challenges was to build a system that could do that heavy lifting and still deliver an accurate location in real time,” Eustice said in a press release. "When you're able to push the processing work to a graphics processing unit, you're using technology that's mass-produced and widely available. It's very powerful, but it's also very cost-effective.”

The pair tested the software on the streets of downtown Ann Arbor. A human was always in control of the vehicle, but in the background, the navigation system made the same set of comparisons it would perform if it were driving autonomously. According to Wolcott and Eustice, it was accurate to within centimeters. They plan to do more testing when Michigan’s Mobility Transformation Facility testing center—which features a private, enclosed city grid—opens this summer.

EVs Move Downmarket at Detroit Auto Show

Carmakers are falling over themselves to announce full-electric, hybrid, and alternative-fuel vehicles at this year’s Detroit auto show. Chevrolet announced a concept for a mid-range fully electric car, the 200-mile-range Bolt, that might cost around $37,500 before tax credits in its 2017 lineup. The Bolt, designed by an Australian subsidiary of GM, will use a battery under development by LG that has a less block-like design than conventional battery packs and could offer carmakers more design flexibility.

The only current widely distributed electric car in that price range, the 84-mile range Nissan Leaf, will see “a lot of enhancements,” in time to compete with the Bolt, says Nissan-Renault CEO Carlos Ghosn. Other companies, such as BMW and Audi, are also commercializing full EVs, but the Chevrolet announcement is a sign that mainstream American carmakers are now stepping into the fray.

Read More

NASA and Nissan Chase Self-Driving Car Technology

Google’s self-driving cars won’t be the only robotic vehicles roaming NASA’s Ames Research Center at Moffett Field in California. The U.S. space agency has teamed up with automaker Nissan to test autonomous driving technologies that could find their way into future vehicles both on the road and in space exploration missions.

NASA hopes the five-year partnership can help improve the autonomous vehicle technologies available for its robotic rovers during Mars missions and other future space exploration. On Earth, Nissan has set a 2020 goal for the market debut of cars that can navigate without human intervention under most driving conditions. Researchers from both organizations aim to begin testing the first of a fleet of self-driving vehicles before the end of 2015.

The work of NASA and Nissan—with one directed to space and the other directed to earth—is connected by similar challenges,” said Carlos Ghosn, president and CEO of Nissan Motor Co, in an 8 January press release. “The partnership will accelerate Nissan's development of safe, secure and reliable autonomous drive technology that we will progressively introduce to consumers beginning in 2016 up to 2020.”

The two organizations have cooperated on technological development in the past. For instance, Nissan used NASA’s research on neutral body posture in low-gravity conditions to develop more comfortable car seats. But hardware and software for self-driving cars could prove to be some of the most transformative technologies to reach mainstream acceptance in the coming years.

Ghosn has suggested that Nissan’s introduction of a commercially available self-driving car could even take place as soon as 2018. He mentioned legal considerations rather than technological roadblocks as the biggest potential stumbling block along any timeline. On the other hand, Nissan engineers have emphasized a less firm deadline in order to leave themselves more wiggle room.

Other observers say that, Ghosn’s reassurances notwithstanding, there remains a list of technical and regulatory hurdles that must be cleared before self-driving cars can be expected to make the world’s roads at least as safe as they are with humans in control. The toughest part of the challenge for robotic cars will be dealing with a mix of automated vehicles and ordinary vehicles driven by humans.

As I noted earlier, the “zero-emission,” self-driving vehicles to be tested by Nissan won’t have the run of the place. They’ll share the NASA testing grounds with potential competitors such as Google. Google has already been making use of the NASA Ames Research Center to test its own self-driving vehicle—a two-seat, all-electric prototype that dispenses with the traditional steering wheel and accelerator and brake pedals in favor of just a start and stop button. The Silicon Valley giant hopes to begin tests of its unoccupied self-driving cars on the NASA research campus sometime this year.

Other carmakers are also racing to develop self-driving vehicles. Mercedes-Benz has begun testing its own robocars at an abandoned naval base in Concord, Calif. Meanwhile, Elon Musk has promised that his Tesla electric cars will be able to operate without human assistance for 90 percent of miles driven starting this year.

Car To Human: You Jerk!

Until our ever-smarter cars can finally take the wheel, the main problem will be to get them to cooperate with us—to sense our intentions, exploit our abilities, and anticipate our errors. 

Swedish researchers say they've made a step in that direction by explaining the longstanding problem of why people drive so jerkily. Until now, the consensus was that it had to do with our difficulties in tracking the road; the researchers say that it is better understood as a reaching problem.

Read More

Cars That Think

IEEE Spectrum’s blog about the sensors, software, and systems that are making cars smarter, more entertaining, and ultimately, autonomous.
Contact us:

Senior Editor
Philip E. Ross
New York City
Assistant Editor
Willie D. Jones
New York City
Senior Writer
Evan Ackerman
Berkeley, Calif.
Lucas Laursen

Newsletter Sign Up

Sign up for the Cars That Think newsletter and get biweekly updates, all delivered directly to your inbox.

Load More