Trained dogs are smart enough to find bombs, drugs, people, and the safest way to cross the street — but only with a capable handler nearby. Now a new system developed at Auburn University could turn canines into remotely guided "super dogs" that can take on risky tasks.
"With our system you don’t have to be in eyesight, versus human guides that do have to be within sight," said David M. Bevly, an associate professor of mechanical engineering atAAuburn University who worked onAthe research.
In the past, remote guidance research focused on other animals or relied on invasive implanted electrodes to give commands. Instead, Bevly and his team created an external, real-time navigation system for trained dogs.
They designed a custom harness equipped with GPS, sensors, a processor and a radio modem that connects wirelessly to a computer system. The pack vibrates slightly on the left or right side and emits different tones to direct the dog.
Protei is a fleet of Sailing Drones, developed primarily to collect Oil Spills under Open Hardware licensing by Open_Sailing, randomwalks and V2_.
It’s just a matter of time till we see these over here…
The affection of a certain island nation for all things robotic — fromAhundred foot tall warfighting mecha toAinfantile therapy robots — is well known. It contrasts sharply with the equally entrenched Western fear of automatons, beginning with the very invention of the term "robot," which was coined in aACzech play that debuted in 1921 in which, naturally, the robots eventually rise up and kill their human masters.
How could two cultures come to such fundamentally divergent conclusions about the status and future of the semi-autonomous helpmates whose increasing presence in our lives seems pre-ordained by nearly every sci-fi vision of the future?
Heather Knight,Afounder of the world’s first (non-industrial) robot census, has made the study of robot / human interaction her life’s work. She posits that the difference between Japanese and American attitudes toward robots is rooted in something much older than even the idea of robots: religion. "In Japan… they’re culturally open to robots, on account of animism. They don’t make a distinction between inanimate objects and humans."
Eric Horvitz, a distinguished scientist atAMicrosoft Research, has an equally distinguished receptionist.
After eight years of working with him, she’s developed a feel for his habits and routines. She can predict how long his meetings are likely to last, knows when he can’t be interrupted and senses when he’ll be running late.
Not bad for an avatar.
The Personal Assistant for Scheduling system, a disembodied head floating on a computer screen outside Horvitz’s office in Redmond, Wash., is one of the most advanced artificial intelligence programs in the world. It can understand speech, detect faces and interpret body motion. It analyzes years of Horvitz’s daily routines to find patterns that suggest the appropriate responses to most workplace scenarios.
The 21st century secretary underscores the breakthrough underway in artificial intelligence and what becomes possible as computers learn to, for lack of a better term, perceive. Through what are broadly called natural user interfaces, machines can increasingly make sense of activity in the real world, from touch, motion, images and sounds, as opposed to information entered through keystrokes.AComputers are learning our language, instead of forcing us to learn theirs.
Posted: October 15th, 2010
at 1:23pm by Koookiecrumbles
Categories: "ninja",9th dan,art,blogs,business,computers,crime,development,diy,education,et cetera,ethics,fo' real?,health,home,hood status,internets,life,mnp is for the children,music,myninjaplease,ninjas are everyehere,philosophy,politricks,real life news,robots,science,tele,too good to be true,travel,weaponry,web
Comments: No comments
In September a driverless Audi TTS will speed to the top ofAColorado’s Pikes PeakAat just under 100 kilometers per hourathat’s right, no driver. It is an early step toward a robo-car that can drive itself, perhaps better than you can.
The World Health Organization projects traffic fatalities to be the third leading cause of mortality worldwide by 2020. And drivers themselves are responsible for 73 percent of these deaths. So automakers are looking at ways they could make cars safer by taking driving out of human hands. Self-driving cars could offer other benefits:ATNO, an international research firm based in the Netherlands, says that they could reduce the time lost to traffic jams by up to 50 percent, and reduce CO2 emissions and fuel consumption by 5 percent.
The Pikes Peak run is a joint project of the Stanford University Dynamic Design Laboratory, the Electronics Research LabA(ERL) for the Volkswagen Group (which owns Audi), and software-maker Oracle Corp. The rough, part-gravel road to the top of Pikes Peak is the route of the annual International Hill Climb rally, an annual auto and motorcycle race. The TTS run will demonstrate whether the car can take curves as fast as a human driverawithout driving off a cliff.
Decision making has two dimensions: A decision-made dimension and a time dimension. For example, a chess game. A game is divided into discrete time chunks, and each time-chunk has a decision made. Say for example, on each move, a player has 15 possibilities of moving on average, those 15 exist on a vertical axis. The time-chunk where each decision is made exists on the horizontal dimension.
Chess is a pretty hard processing problem, because the amount of possible decisions from all previous time chunks makes each following decision pretty difficult.
Free dance is similar in that it is divided into time-chunks, and each chunk also has a particular decision to be made. This decision involves the position of the arms, the legs, the feet, the head, etc. There are additional factors, however: the positions have to match the music, the positions have to match what other people around are doing, and the positions have to fit into the space available. Additionally, each position has to take into account the known flexibility and balancing ability of the body being controlled.
And finally, a particular routine has to be changed after a guaged time so as not to get ‘boring’. The set of routines have to be pre-learned, not by actually having the limbs being moved into position, but by simply observing another body making the moves, reproducing the moves and then comparing the reproduction to the original.