Thursday, 16 August 2012

How to Teach a Robot to Improvise

Self-piloted drones have become sophisticated enough to land on moving aircraft carriers, but put a single unexpected tree in the way, and they will crash. Now a five-university group that includes specialists in biology, computer vision and robotics is trying to teach drones to dodge obstacles on the fly. Working with $7.5 million from the Office of Naval Research, the scientists aim to build an autonomous, fixed-wing surveillance drone that can navigate through an unfamiliar city or forest at 35 miles an hour.

Video: Disney's 3-D Facial Scruff Technology Reconstructs Beards Down to the Individual Hair

Disney's 3-D Facial Hair Reconstruction Technology via Disney Research
Face capture technology has come a long way, especially as 3-D stereoscopic imaging and the like have made leaps forward in recent years. It’s now relatively easy to capture a face in 3-D and reconstruct it digitally for applications such as the amazing CGI you see in movies like The Avengers (Ruffalo-Hulk was pretty visually awesome, no?). But facial hair is another story altogether. Current face capture systems don’t capture it well, and the skin that it obscures on the face then becomes an issue as well. Disney Research is changing that.

With New Programming, Autonomous Airplanes Can Navigate on the Fly With No GPS

Self-Navigating Drone MIT
Self-piloted drones may be able to land or fly almost anywhere -- even aircraft carriers -- but they need some complex navigation skills to do it, including the somewhat existential ability to know where they are in the world. But this is difficult

Creating 150 MPH Hurricanes in a Giant Aquarium in Florida


SUSTAIN Cambridge Seven
About two years ago, Brian Haus, the chair of the Division of Applied Marine Physics at the University of Miami, was studying storms in the western Pacific ocean, off the coast of Taiwan. He and his team are chasing hurricanes. Sometimes the hurricanes completely miss the sensor-packed buoys placed in their path to track power and speed. Sometimes they don't.

How Tablets Will Remake Television



Double Vision Alison Seiffer
The days of leaning back to watch TV have ended. Eighty-eight percent of tablet owners say they use the device in front of the tube; they find tweets, news, video and other information related to the program they’re watching. Afraid of losing eyeballs, networks have released dozens of one-off apps with additional programming

Karate Experts' Superhuman Punch Comes From a Unique Brain Structure

Karate WC Tampere 2006 Wikimedia Commons
If you've seen the board-breaking power of a professional martial artist and thought it looked superhuman, don't worry: for a while now science couldn't fully explain it, either. The punches delivered by a top-notch fighter are so tough that

Wednesday, 15 August 2012

Computer Models Help Predict Wildfire Behavior


U. S. Fish and Wildlife Service - Northeast Region/CC BY 2.0
The Top Park bonfire in Colorado endure ages covered both asperous breadth and busy areas, affective in a decidedly aberrant way, at atomic partially because of the accident already done by case beetles in the area. The consistent asleep copse actualize added ammunition added backwoods fires and change the backwoods canopy, which makes it added difficult to action the fires.
A new computer clay software alleged HIGRAD/FIRETEC has been developed through a affiliation amid the US Backwoods Service Rocky Mountain Analysis Station and Los Alamos National Laboratory (LANL) to advice adumbrate bonfire behavior in case beetle-ravaged areas.
LANL explains, "HIGRAD/FIRETEC is a physics-based, 3-D computer cipher advised to simulate the consistently changing, alternate accord amid blaze and its environment. It does so by apery the interactions a part of fire, fuels, atmosphere and cartography on a mural scale. HIGRAD is a computational fluid-dynamics archetypal that represents airflow and its adjustments to terrain, altered types of ammunition (vegetation) and the blaze itself. FIRETEC combines physics models that represent combustion, calefaction transfer, aerodynamic annoyance and turbulence."
The software models blaze behavior application abstracts that advisers accept calm in forests area case beetles accept done their damage. Their analysis is absorption appropriate now on what blaze behaviors are acquired by the access in asleep ammunition and what behaviors are acquired by the changes in the backwoods awning which acquiesce top apprehension to advance the blaze added easily.
More importantly, HIGRAD/FIRETEC simulates the activating processes that action aural a blaze and the way those processes augment off and adapt anniversary other. So, the models appearance how if a blaze hits a application of asleep copse it accelerates tree-top fires that can absorb reside trees. Previous models accept affected a compatible damp agreeable throughout the forest, authoritative them abortive for forests area case beetles accept been
By addition these differences out, firefighters will be bigger able to action these added alarming and aberrant fires.