The search giant’s virtual assistant—called Assistant—is more ambitious than Siri and Alexa.
from Technology Review Feed – Tech Review Top Stories http://ift.tt/2dHIoVw
via IFTTT
GM and the US Army have developed a monster hydrogen fuel cell truck
-
The Chevy Colorado ZH2.
GM -
That’s a stretched chassis and 37-inch wheels.
GM -
The car’s hydrogen fuel cell produces water as its only byproduct. Excellent for maintaining the integrity of your bodily fluids.
GM
Late last year, we got news that General Motors would work with the US Army to develop a hydrogen fuel cell-powered pickup truck. On Monday at an annual US Army association meeting, GM took the wraps off the beast, which the Army will test in Michigan over the next year.
The US Army’s tank research center collaborated with GM to build the Chevy Colorado ZH2, which has a reinforced body that’s six-and-a-half feet tall and seven feet wide. The truck will chew up terrain with 37-inch tires and a special suspension built for off-road handling.
The ZH2 has a single motor that’s powered by a hydrogen fuel cell and a battery. The advantage of that hydrogen fuel cell is that the only byproduct is water, and the electricity-powered engine is quieter than a traditional combustion engine. It also gives off less heat, which GM said would help the car in stealth situations, where the Army would want to reduce acoustic and thermal signatures. According to Wired, the ZH2’s hydrogen fuel cell produces two gallons of water an hour.
The truck also comes equipped with what GM called an “Exportable Power Take-Off unit (EPTO)â€â€”basically a 25kW battery that’s charged by the fuel cell and can be removed from the truck to power anything else.
Needless to say, the ZH2 won’t be found at your local dealership anytime soon. This is a military research vehicle through and through.
The Army is leasing the truck from GM for a year to test at proving grounds in Michigan, where it will be evaluated for potential use in combat situations. By the end of 2017, the Army hopes to have a full picture of how well the truck performs with respect to wheel torque, fuel consumption, and water-byproduct quality.
Of course, the issue with hydrogen fuel cells has been the same for decades—although fuel cell vehicles take minutes to refuel unlike battery-powered cars, hydrogen can be difficult to store without a high pressure container or very cold temperatures. Wired notes that this may not be such a problem for the Army, as it could repurpose JP-8 jet fuel supply tankers to supply H2, or it could make hydrogen from the jet fuel itself.
Clearly, hydrogen supply issues aren’t stopping GM from building research vehicles. The Colorado ZH2 follows GM’s announcement of an unmanned undersea vehicle (UUV), which it built in partnership with the US Navy. That UUV is currently being tested in a pool for “weeks if not months of endurance†in underwater environments.
On top of that, GM said it had already logged 3.1 million miles of driving between 119 hydrogen fuel cell vehicles. (In fact, GM even built the world’s first hydrogen fuel cell vehicle, called the Electrovan, back in the 1960s.)
Although this truck is for research purposes only, the Army’s testing could help hydrogen fuel cell vehicles make their way into the general market. Charlie Freese, GM’s executive director of Global Fuel Cell Activities, said in a statement that the Detroit automaker would benefit from the Army’s research. “Over the next year, we expect to learn from the Army the limits of what a fuel cell propulsion system can do when really put to the test.â€
Listing image by GM
from Ars Technica http://ift.tt/2dqFvdi
via IFTTT
This Gene Mutation May Make You Crave More Greasy Food (And Shrink Your Sweet Tooth)
We all love eating junk food even though we know it’s not part of a healthy diet. Some of us prefer burgers and fries that widen our waistlines, while others struggle with diabetes-inducing cookies and candies. Either vice can lead to eating more calories than we need and can contribute to weight gain. But scientists were never really sure how the body regulates our food preferences. Now a team of researchers may have discovered a gene that acts as a switch, dictating whether we like high-fat or high-sugar foods more. Â
In a study published online today in the journal Nature Communications, scientists at the University of Cambridge found that neuronal pathways involving the melanocortin-4 receptor (MC4R) in the brain play a critical role in influencing food choice. "Our work shows that even if you tightly control the appearance and taste of food, our brains can detect the nutrient content," said senior study author Sadaf Farooqi, a professor of Metabolism and Medicine at the University of Cambridge.
Previous studies in mice had shown that gene mutations that disrupted the MC4R pathway resulted in the mice eating a lot more fat. But when these mice were presented with sweetened food and water, they didn’t like those options and ate a lot less. Â
Farooqi and her team decided to test if the receptor functioned the same way in humans. The researchers recruited 14 people with rare MC4R mutations that led to decreased activity melanocortin signaling, and compared them with lean and obese individuals. Participants were given three options of chicken curry that looked and tasted the same but contained either low, medium or high levels of fat. Next, the participants were presented with another all-you-can eat buffet–this time of a dessert with varying levels of sugar.
Although they couldn’t tell the differences in fat content or sugar content in the meals, people with MC4R mutations ate almost double the amount of high fat than lean individuals did (95 percent more) and 65 percent more than obese individuals. But interestingly, they liked the high sugar dessert less than their lean and obese counterparts and ate significantly less of all three desserts compared to the other two groups.
The results are exciting because they are the first to point to a genetic preference for certain kinds of foods, says Claudia Doege, who studies obesity at Columbia University. “We know that 40 to 60 percent of obesity is inherited but it has been very difficult to find which genes drives these cases,†she said.
That doesn’t mean it’s okay to have a daily dose of hamburgers or pastries. As with all biological mechanisms, the chemicals and receptors that regulate appetite or food preference form a complex web of checks and balances, according to Doege. But understanding the role that the MC4R pathway plays in the brain, could help develop new therapies for treating metabolic disorders like obesity.
from Popular Science – New Technology, Science News, The Future Now http://ift.tt/2dFQ5eb
via IFTTT
Google Builds a Robotic Hive-Mind Kindergarten
How many robots does it take to open a door? If the robots are trying to figure out how to do the task from scratch, then it helps to have as many as possible involved.Â
In three separate research papers posted online Monday, researchers at Google and other Alphabet subsidiaries showed several ways in which robots can learn to perform simple tasks more quickly by sharing different types of learning experiences.Â
The researchers are training teams of industrial robots to perform simple tasks using a technique called reinforcement learning, which combines trial and error with positive feedback. For the moment, these tasks are extremely simple, like opening doors or nudging objects around. But such advances will be crucial if robots are ever going to be capable of helping with everyday chores like folding the laundry or doing the dishes.
Although robots are becoming cheaper and more capable, programming them to behave reliably in unpredictable everyday situations is a near impossible task. Reinforcement learning offers a solution, by letting robots essentially program themselves as they learn on the job. But it can be very time-consuming for an individual bot to try many different ways of performing a chore. Sharing the learning process, a technique often called cloud robotics, can help accelerate the process, although the idea remains at an early stage (see “10 Breakthrough Technologies: Robots That Teach Each Otherâ€). Â
In the three papers released Monday, Sergey Levine, a research scientist at Google who is leading the robot learning effort, and colleagues detail several learning strategies that can be distributed across a group of robots.
In each case, the robots involved use neural networks that try to predict the result of different actions. Each robot varies its behavior slightly, and then reinforces the variations that give bigger rewards. These networks are then periodically fed back to a central server which builds a new neural network that combines all of the learned behavior, and that network is redistributed back to the robots for another round of training.
In the first experiment, the goal was turning a door handle and opening a door, and four different robots were set to work practicing on different doors and handle types. “Since the robots were trained on doors that look different from each other, the final policy succeeds on a door with a handle that none of the robots had seen before,” Levine wrote in a blog post coauthored with Timothy Lillicrap, a researcher at Google DeepMind, and Mrinal Kalakrishnan, a researcher at X, Google’s “moonshot” research facility.
In the second experiment, the robots’ learning process was sped up thanks to the interactions of a person who guides a robot arm. And in a third, a robot figured out how to move and rotate objects using input from a camera and a learned ability to predict how actions would change the picture—what the researchers describe as a simple physical model of the world. Stephanie Tellex, an assistant professor at Brown University who studies robot learning, says this is an exciting idea. “Predicting the physical effects of actions like pushing is exciting because it enables the robot to understand something about how the world works,” she says.
The company is evidently keen to make the most of what could be a coming revolution in the field thanks to the application of machine-learning techniques. Already some robotics manufacturers are exploring ways to use reinforcement learning to streamline the programming of their products.
“Of course, the kinds of behaviors that robots today can learn are still quite limited,” the authors wrote. “However, as algorithms improve and robots are deployed more widely, their ability to share and pool their experiences could be instrumental for enabling them to assist us in our daily lives.”
from Technology Review Feed – Tech Review Top Stories http://ift.tt/2d09OrF
via IFTTT
Dazzling Long Exposures Capture the Fireflies of Japan
When you see a firefly, it’s only for a moment. The bright light blinks and vanishes until it magically appears a few feet away. But photographer Kei Nomiyama freezes the dance with long exposures that make hundreds of fireflies appear frozen in mid-air.
Nomiyama is an environmental science professor, but loves to spend his free time photographing the world he studies. “I became a scientist to protect nature, and I have an interest in photography to record nature,†he says. The fireflies thrive in the forests of Shikoku Island where Nomiyama lives, and he’s spent the last eight years documenting their mating ritual with his camera.
The fireflies are most abundant during Japan’s rainy season between May and June, where they live a brief but beautiful two-week adult life. During that period, Nomiyama makes frequent into the forests around central Shikoku Island, seeking the perfect patch of trees or river for his shoot. Once he finds a location, Nomiyama makes long exposures up to 30 minutes with his Canon EOS 5D Mark III and Sony Alpha a7R II. Later, he digitally composites multiple frames together.
The final images are overflowing with hundreds of tiny lights. In the early 20th century, firefly hunters captured thousands of the insects to illuminate hotels and private gardens in Tokyo. Nomiyama just needs his camera.
from Wired Top Stories http://ift.tt/2dTD4lD
via IFTTT
The Human Cost of the Lithium Ion Revolution
If you have a cell phone, laptop, a hybrid car or an electric vehicle, you may want to sit down. This is going to hurt.
You have probably heard of blood diamonds and conflict minerals, maybe you’ve even read up a bit on how big consumer tech companies are trying (and, in some cases, being forced by governments) to sort out where the materials they use for their gadgets come from. But stories about “supply chains,†“globalization,†and “poor working conditions†can a world away, or just plain academic.
In a sweeping, heartbreaking series, the Washington Post is putting the lie to that.
Take the example of Yu Yuan, a farmer who lives near a graphite factory in northeastern China. In a video, he swipes at shimmering grime accumulated in his window sill and points at a barren cornfield. The crops turn black with graphite dust he says, and don’t grow properly. Hh and his wife worry about the air they’re breathing, and their water is undrinkable, polluted by chemicals dumped from the graphite plant. “There is nothing here once the factory is done damaging this place,†he says.
Over two pieces so far, the Post has traced the path of first cobalt and then graphite as they make their way from mines to factories and ultimately into our hands as the cathodes and anodes, respectively, for lithium ion batteries. Each of them is a remarkable blend of globe-spanning investigative journalism, business reporting, and an appeal to us to confront the consequences of owning the devices we surround ourselves with.
While graphite is mined and processed mostly in China, a huge amount of cobalt comes from mines in the Democratic Republic of the Congo, where “artisanal†miners sometimes dig through the floor of their own houses. Mines collapse frequently-injuries and death are commonplace.
The raw materials stay or end up in Asia, where companies you’ve probably never heard of turn them into battery parts. There the largest battery makers in the world, including Samsung SDI, LG Chem, and Panasonic, purchase the components and turn them into batteries that go into phones, computers, and cars the world over.
Lithium batteries are prized for being light and having a high energy density compared to other battery chemistries that came before. The modern smartphone would be difficult to imagine without a lithium battery as its power supply. They help power hybrid cars and the small, but fast-growing fleet of all-electric vehicles wouldn’t exist without them.
Interest in electric cars, in particular, is fueled by claims that the vehicles are cleaner and better for the environment. That may be true in the countries where they are mostly sold. But when we consider the bigger picture, the reality is something else altogether.
(Read more: The Washington Post, "Why We Still Don’t Have Better Batteries")
from Technology Review Feed – Tech Review Top Stories http://ift.tt/2dnOagP
via IFTTT