Archive for the ‘Technology’ Category
Army Developing Drones That Can Recognize Your Face From a Distance
And even recognize your intentions
Of a handful of contracts just handed out by the Army, two are notable for their unique ISR capabilities. One would arm drones with facial recognition software that can remember faces so targets can’t disappear into crowds. The other sounds far more unsettling: a human behavior engine capable of stacking informant info against intelligence data against other evidence to predict a person’s intent. That’s right: the act of determining whether you are friend or foe could be turned over to the machines.
That’s a bit disquieting whether you are an insurgent warfighter or not. But back to the overarching topic at hand: The U.S. military is pulling in more ISR data than it knows what to do with these days, a lot of it useless noise that’s inconsequential to ongoing operations. And, as DR notes, the strategy in Afghanistan has changed from one of winning hearts and minds through nation building projects to targeting specific bad guys.
The hard part is keeping up with the bad guys, and that’s where Progeny Systems Corporation’s “Long Range, Non-cooperative, Biometric Tagging, Tracking and Location” system comes into play. The facial recognition layer of its technology is pretty standard: take some 2-D pictures of a target’s face, use them to build a 3-D model, and then use that 3-D model to recognize the face later.
But that’s not necessarily easy. It’s difficult enough for computers to pull off biometric facial recognition when the subject is stationary and looking straight at the camera. Toss in the many variables inherent in aerial ISR--a moving target who may be in profile or looking downward, a moving drone, low resolution cameras, etc.--and it’s a major challenge.
Progeny’s system, if it works the way the company and the Army envision it, needs just 50 pixels between the target’s eyes in a 2-D image to build the 3-D model. “Any pose, any expression, any face,” the company’s lead biometric researcher tells . From that model stored in Progeny’s database, the system could identify the target from an even lower resolution image or video.
The closer the drone is to the subject, the better all of this works. But progeny also layers in a second kind of recognition that can work at more than 750 feet. This “soft biometric” system basically takes in a bunch of non-facial but otherwise outwardly relevant data--skin color, height and build, age, gender--to build a larger kind of model for its vision algorithms to work with. If a body is moving through the crowd, Progeny claims that a drone circling high overhead can keep track of him or her simply using this larger, whole-body identification system.
But what good is tracking if you don’t know who your enemies are? Another contract handed out to Charles River Analytics seeks to develop a human behavior engine known as Adversary Behavior Acquisition, Collection, Understanding, and Summarization (ABACUS). It mashes up all kinds of behavioral data into a system that churns out an assessment of adversarial intent, determining if a subject has enough built up resentment toward the U.S. and its aims to be a potential threat.
So pretty soon the drones may know who you are, where you’re going, and what you’re planning to do when you get there.
[]
Inside the Factory: How a Chef’s Knife Is Made
PopSci goes to Germany to witness the cutting edge of manufacturing
The raw material comes into the factory on huge spools of sheet steel, each sheet the thickness of a knife. The steel is cut into individual blanks, destined to become individual knives. About three weeks elapse between when a blank comes off the spool and when it emerges, a finished knife, at the other end.
In my breathless tour of the factory, I watched as a giant press cut and stacked the blanks, which are made of the company's secret blend of stainless "special formula steel." The blank is transported to another building, where the first of the factory's 90-odd industrial robot arms takes it in hand.
In the classic design, the knife has a thickening where the blade meets the handle (aka the bolster). This is formed first, by heating the middle of the blank, and then pressing the metal's two ends together so the molten middle bulges and widens, in a process that my contact specifies is called upset forging. Next, a drop forge shapes the bolster, before the blank is quickly cropped into the rough shape of the knife it's going to be.
After that, it proceeds through a series of cooling, supercooling, and heat-tempering steps that give it its corrosion resistance and toughness. This is one of the benefits of the special steel, I'm told -- it heats and cools in very predictable ways, allowing the factory to use more precise temperatures rather than temperature ranges.
After the tempering, any distortions or warpings that the heat has created in the blade are hammered out by a highly skilled human, who picks up and eyes each knife, one at a time, and flattens any that need flattening with precise strokes of a little hammer.
The knife passes into the hands of another series of robots, which use grinding wheels to narrow down the thickish blank into the tapered contour of a blade. Only roughly, though -- the fine grinding and sharpening, as well as putting on the handles, is left to the factory's humans, who wear puffy gray overalls and exude the confidence that comes with being extremely good at your job, and quite possibly coming from a line of knifemakers generations old. Wooden handles are glued onto the tang of the knife and then riveted in place; plastic handles are simply melted on by heating the tang and inserting it in a ready handle. The edges of the handles are smoothed by robots.
Finally the knife is cleaned and passes onward to the scrutiny of the quality assurance women. If it has no flaws -- there's a big photo-book of possible flaws -- it gets packed up and winds up in someone's kitchen.
Check out the of pictures from the knifemaking process.
brightcove.createExperiences();
Kepler Analysis Projects One-Third of Sun-Like Stars Have an Earth-Like Planet Orbiting
If that turns out to be the case, that’s big news of course. The habitable zone, or the “goldilocks zone” as it’s often known (not to close to the star, not too far away), is the orbital range where it’s possible for liquid water to exist. Thus, it’s the range where life as we know it could feasibly take root.
The planet-hunting Kepler observatory is designed specifically to seek out planets orbiting distant stars, and thus far its been a boon for exoplanetary science. In 136 days it has scanned some 150,000 target stars looking for the signature wobble exerted on those stars by orbiting satellites. In doing so, it has found 1,235 potential planets.
It’s from that data that Traub has extracted his conclusion. He looked particularly at the stars that are most like our sun--those classified F, G, or K. He then looked at the kinds of planets that are most often found orbiting them and at what ranges they orbit. In his analysis, he notes many interesting (and somewhat expected) things, like the fact that nearly a third of the planets Kepler has found orbit their stars in less than 42 days, putting them too close to be in the habitable zone (this is also because planets closer to their stars are easier for Kepler to see).
Larger terrestrial planets out there in the habitable zone are harder for Kepler to spot, but that doesn’t mean they’re not there. And Traub says his number crunching allows us to get a pretty good idea of how many there should be. Using some math we don’t pretend to understand, he plugged in the numbers for longer orbits--orbits in the habitable zone--into his analysis. The finding: "About one-third of FGK stars are predicted to have at least one terrestrial, habitable-zone planet."
That’s not to say they are inhabited, or that they do have liquid water, or that they even exist. But Traub’s math suggests that they should exist, at least until more data changes the equation. And for now, that spells a lot of potential goldilocks planets. Read the full paper via .
[]
NASA’s Falling UARS Satellite Found in Remote South Pacific
Maybe next time, Canada
See our gallery of the space race's greatest falls to Earth. (List compiled by .)
NASA has been tracking UARS for some time now as the decommissioned satellite’s orbit has been decaying. Much of the satellite was expected to burn up on re-entry, but experts estimated that roughly two-dozen pieces of the massive satellite would survive and could potentially be a threat to people or objects on the ground. Given UARS’s speed and the many variables involved (this is a decommissioned satellite, after all, so re-entry was completely uncontrolled) there was no telling exactly when or where UARS might land.
On Saturday, when the final descent began, previous calculations had placed the crash window across a large swath of northwestern North America. The Internet rumor machine fired up and sightings across Canada and the Pacific Northwest proliferated. But by that point updated U.S. Air Force calculations placed the satellite thousands of miles away in another hemisphere, and NASA has confirmed those calculations. UARS is now resting peacefully in the South Pacific, somewhere southwest of Christmas Island were small islands are scattered across a lot of water.
The difference between Seattle and Samoa? Just a few minutes. NASA said UARS came in for its rough landing several minutes earlier than they had projected. What they won’t say is how they know this--they referred those questions to the USAF, which also isn’t talking. Were DoD missile tracking assets employed in tracking UARS? The Air Force would rather not say at this point, but one would think something like this would be good practice.
UARS is to come crashing back to Earth, and it won’t be the last. In late October or early November a German astronomy satellite will make its uncontrolled final plunge back to Earth. Though smaller than UARS, more pieces are expected to survive re-entry (a total of 30 are expected, possibly including sharp pieces of mirror). Let’s hope that one finds a nice stretch of uninhabited ocean as well.
[]
BullDog: A Bigger, Scarier Version of BigDog Gets Closer to the Battlefield
Boston Dynamics is building a bigger, sturdier version of the military’s future trusty companion, and will likely unveil it within a few months. The company’s founder and president, Marc Raibert, shared the LS3 robot's progress Tuesday at a keynote speech at the 2011 IEEE International Conference on Intelligent Robots and Systems. Apparently LS3 (Legged Squad Support System) has been nicknamed BullDog, .
Alas, no fun video yet, as Boston Dynamics is apparently waiting for permission from DARPA to release it.
BullDog, like BigDog, is designed to carry hundreds of pounds of gear for armed forces, ambling over rough terrain and following humans without complaint. The larger version will carry 400 pounds, last 24 hours and carry enough fuel for a 20-mile trek. It will also be able to jump over obstacles, and more easily regain its footing after it falls over. BullDog will also have greater navigational autonomy than BigDog, IEEE says.
The most significant change may be that it’s significantly quieter than BigDog, which is quite obnoxiously, buzzingly loud:
Granted, a prancing, unstoppable four-legged metal beast probably doesn’t need stealth to look awesome and surprise the enemy.
BullDog is a 30-month, $32 million project funded by DARPA’s Tactical Technology Office and the U.S. Marine Corps Warfighting Lab. The project started in early 2010, so we anticipate a full unveiling sometime next year.
Until then, content yourselves with some of .
[]
The Sunspots That Kicked Off This Week’s Solar Storm May be Just Warming Up
AR1302 unleashed a massive coronal mass ejection on Saturday that struck a glancing blow off Earth’s atmosphere yesterday, triggering brilliant auroras across the Northern Hemisphere. So far, the storm hasn’t caused any serious trouble here on the ground. Saturday’s solar explosion didn’t connect with a direct hit, and it is expected to do nothing more than continue to provide electrifying light shows to sky-gazers in Europe and Asia this evening.
But AR1302 is also not slowing down, and as the week wears on it will turn to face Earth more directly. An SWPC bulletin yesterday warned that for the next 3-5 days, we’re squarely in the solar storm’s sights. Another blast like Saturday’s and we may feel it here on Earth in the form of disrupted communications. A larger blast could do even more damage to the power grid and other infrastructure.
Just another thrilling week in the buildup to 2013’s solar maximum. See the sun as NOAA’s GOES-15 sees it today below.
[, ]