Automark

A poll released today suggests that driverless cars appeal most in China and India and least in Japan, with English-speaking countries—the only comparison group—taking the middle ground.

The Japanese position at the bottom and India’s near the top are strange. Could Japan’s reputation for robo-philism be unjustified? Was the survey—conducted online—unrepresentative of opinion in China and India?

In their paper on the survey, authors Brandon Schoettle and Michael Sivak, of the University of Michigan, address the second question. They argue that “though the respondents in these two countries may not be representative of the overall population, they are likely to be representative of those individuals who would comprise the initial market for autonomous and self-driving vehicles in these countries.”

Here are some key numbers. Respondents who were “very interested” in having a totally self-driving car peaked at 47 percent in India (a little ahead of China) and cratered at 8.5 percent in Japan. The median respondent (that is, the one at the 50th percentile) was prepared to pay US $1600 extra for such a car in China; the sum was just a tenth as much in India, and precisely zero in the other four countries.

To probe the minds of possible early adopters, the pollsters list the premium that the 75th percentile would pay: $8,000 in China and $2,350 in Australia, with the U.S., Britain and India coming in a few hundred dollars lower. In Japan, these 75th percentilers would pay a paltry $465.

So how to explain Japan’s apparent coolness to autonomous vehicles? Maybe its reputation for liking robots has been exaggerated. Maybe one’s attitude to robots is irrelevent to the decision to buy them. Or maybe the Japanese are simply further along the famous Gartner hype cycle than we are; maybe they’re standing where the rest of the world will be, soon enough.

via IEEE

Sharp eyesight can’t reliably keep you vertical if you have lost your sense of balance, which shows just how important it is to have several different senses at your disposal.

That’s the idea behind a system recently proposed by two Korean engineers to make a car better at finding itself a parking spot. Rather than depend on ultrasound sensors mounted on the grille, as some parking programs do, or on cameras mounted on all sides, the researchers fuse the two kinds of sensor. Three, if you count data from the odometer, which measures the movement of the car.

The auto industry is working on ways to automate the entire parking process, not just the last bit. And Audi, Volvo and Nissan have all shown off parking-space finders that perform well in controlled circumstances, say by linking to a parking lot’s WiFi system. But to work alone in an uncooperative world, cars will have to wring more data from their existing sensors.

Jae Kyu Suhr, an IEEE member, and Ho Gi Jung, a senior member, recently laid out a way to do that. The researchers are affiliated with Hanyang University, in Seoul; their work was supported by the Hyundai Motor Company.

Sensor fusion is tricky because the various sensors look at an object from different angles. The researchers solve the problem with a lot of math and some plain old ingenuity.

First, the car moves past a parking area, scanning for the marked edges of parking slots and for obstacles (such as a parked vehicle). The several cameras and the two ultrasound sensors of course provide different vantage points—what is visible to one sensor may be obscure to another.

At a given point, called a frame, the car processes all the information available to classify the parking area according to its structure—an array of rectangles laid out orthogonally, or on a bias, or in a staggered fashion, and so on. Now that the car knows what it’s dealing with, it knows what to look for: say, for the characteristic edges of a staggered rectangular parking spot.

By breaking down the job in this hierarchical fashion, the researchers say, their system can keep the computational time to just 32 milliseconds, compared with 82 ms for a vision-only system. “These results reveal that the proposed system can surely operate in real time,” they write.

Next, the car moves on to the next vantage point, giving each sensor system the chance to get a better view, at least of some features. Because the odometer tracks the car’s position, the system can figure out the new angle of observation and use it to update its earlier estimate. Besides improving accuracy, this continuous updating helps the system handle roads that aren’t perfectly flat.

Finally, the system offers a selection of possible spots to the driver, who selects one by punching a touchscreen. From this point, the system works like existing car-parking programs.

There are a few drawbacks. The system can’t work as billed at night or in dimly lit, reverberating underground parking lots. To overcome that problem the researchers are developing new algorithms and working with specialized equipment, such as cameras with greater range. Rather than try to make one size fit, all, they’d have independent programs to work in the light of day, at night, and in closed spaces.

Humans are still a factor in the adaptation of automatic braking

The remarkable thing about letting a car do the braking for you is not that the car stops. It’s how late the car hits the brakes. It’s almost as if a teenager were testing his or her reflexes. Those of us raised on automatic transmissions and cruise control may expect cars to take flighty human drivers out of the loop rather quickly. But if my ride in a test vehicle at the 2013 Frankfurt Motor Show is any indicator, carmakers are taking their time taking over. Even the most imperturbable driving instructor might get jumpy using today’s autonomous emergency braking (AEB), also called advanced emergency braking systems.

AEB isn’t even a teenager: Mercedes-Benz introduced an early version with its 2005 S-Class. That system used radar to detect obstacles, warned drivers, and primed brakes so that they would be more effective when the driver finally used them. Yet in an indoor test in simulated foggy conditions, the car’s radar failed to activate the system. A journalist crashed one into another Mercedes-Benz during a televised demonstration. Company engineers later decided that the garage’s steel interior had confused the radar. But the technology is maturing, and the European New Car Assessment Programme (Euro NCAP), a public-private car-testing body and the counterpart to the U.S. NCAP, will require AEB to obtain its highest safety rating next year

Now the state of the art is to use more than one type of sensor to cross-check for obstacles, carmakers say. Some complement radar systems with optical cameras. Today’s S-Class has short-range and long-range radar, optical cameras, and ultrasonic detectors for the closest obstacles. Optical cameras can be fooled by sunlight, wet roads, and night, of course, and ultrasonic sensors work only at the shortest ranges and lowest speeds. Certain research vehicles also include lidar, a radarlike system that uses light rather than radio waves. As the instruments grow smaller and cheaper, carmakers may include lidar in production cars as well.

Yet carmakers still hesitate to override a driver’s instincts. The German auto club ADAC reported in a test [pdf] that it deducted points from a BMW 5 Series for initiating only partial braking after warning the driver. But such a limited action may please self-confident drivers. Harald Barth, a product marketing manager at car supplier Valeo, says that one reason carmakers have kept the brakes on autonomous driving is that they want to win the trust of drivers. “We need not just to offer good systems but also to educate the end user. We are going step by step,” he says.

That will also give engineers more time to figure out how humans react to having control taken away from them. Last year, a pair of studies applied analyses called system-of-systems and operator sequence diagrams to AEB scenarios. They both found that when autonomous systems attempt to take over from human drivers, humans do not always respond well. Or sometimes drivers respond too well and do not react in time to take over again when the autonomous systems attempt to return control to them. The latter study sounded a grim note: “There are no formal methods for testing the performance of AEBs from either a technical or human factors point of view. The effectiveness of AEBs will, however, become increasingly clear in the coming years through fatality and injury statistics,” its authors wrote.

Going slow will also give Euro NCAP and other testing bodies more time to improve their testing capacity. Now Euro NCAP uses a small trailer as the crash target, allowing only simulated rear-end collisions, but it says it will develop targets simulating pedestrians, among other improvements [pdf]. For pedestrians, if not anxious driving instructors, that should be a relief.

via spectrum.ieee