<![CDATA[io9: mad robotics]]> http://cache.gawker.com/assets/base/img/thumbs140x140/io9.com.png <![CDATA[io9: mad robotics]]> http://io9.com/tag/mad robotics http://io9.com/tag/mad robotics <![CDATA[How To Train Your Robot To Recognize You]]> When you come home from work at night, does your robot greet you at the door expectantly, or does it sit there impassively in its recharging node because it can't tell the difference between you, the mailman, and Emilio Estevez? Today's computer scientists are hard at work making sure tomorrow's robots won't leave you feeling emotionally shunned. Check out how researchers at the Carnegie Mellon University Robotics Institute are using something called "Active Appearance Modeling" to improve face recognition algorithms that will make your bot snap to attention when it sees your face.


Recognizing a face is harder than it sounds. Using Active Appearance Modeling (one of the common methods in use today), a computer has to compare a face it sees to an "average face" it has previously learned. It works pretty well when the subject smiles and stares right into the computer's camera, but in real life, lighting, facial expression and "3D pose variation" present serious obstacles.

The Robotics Institute team is working on that last bit. Whenever you turn your head, part of your face is occluded. Without the right features to make its comparison, the computer can't recognize you. New algorithms and programming methods allow for the creation of 3D face meshes that can be adjusted on the fly to fit the subject's face, even if she turns partly away from the camera.

Of course, the government will use this technology to track our every move long before we have friendly helper robots who know us on sight, but it's nice to know we live in a world where something called the Robotics Institute actually exists. Image by: Robotics Institute.

AAM Fitting Algorithms. [Carnegie Mellon University Robotics Institute]

]]>
http://io9.com/390636/how-to-train-your-robot-to-recognize-you http://io9.com/390636/how-to-train-your-robot-to-recognize-you Thu, 15 May 2008 08:00:00 PDT Ed Grabianowski http://io9.com/index.php?op=postcommentfeed&postId=390636&view=rss&microfeed=true
<![CDATA[Suburb-Eating Robots Run on Fat Reclaimed from Liposuction]]> It may look like a smiling mecha puppy of vast proportions, but this suburb-eating robot is a vicious destroyer of suburbs and suburbanites whose giant legs pulverize housing tracts in order to plant new forests. The creation of Australian firm Andrew Maynard Architects, the suburb-eating robots will be deployed to clean up abandoned, decaying suburbs in Australia when peak oil forces people to stop driving cars and move into the urban centers. These mega-bots are going into production in 2019, and their engines will be fueled entirely by human fat.

The folks at Andrew Maynard Architects explain their project, with tongues planted mostly in cheek:

The age of the outer-suburb is soon to come to an end. Many analysts believe that Peak-Oil will be reached soon after 2011. When we hit Peak-Oil we will not only have no petrol to run cars, furthermore we will no longer have many of the goods we need and there will be huge food shortages as food production and distribution relies heavily on oil based fertilisers which drastically increase yields.

With no cars people will no longer be able to reach the suburbs and hence metropolitan populations will swell as suburban refugees are forced to wander into the cities . . . The suburbs will decay . . . At Andrew Maynard Architects, we have decided to give mother-nature a hand. We have begun designing the first suburb eating robot and we hope to go into production in early 2019. We have called our robot the CV08. In short, CV08 consumes the abandoned suburbs through its front 2 legs. It processes the materials and fires off compacted recycling missiles to awaiting recycling plants. CV08's middle legs and one rear leg follow the front legs to terra-form the newly revealed earth with native Flora and Fauna. Vast stocks of the Flora and Fauna are stored within CV08 in carbonite sleep until they are required to colonise what was previously suburban wasteland.

Over 50% of Australians are currently over-weight due to complete car dependence, a sedentary lifestyle and over eating. With this in mind the 6th leg has been designed to pick up, and apply liposuction to over-weight Australians that have been to slow and unfit to migrate into the denser areas with the rest of the population. As there is no longer a steady stream of oil, CV08 fuels itself with the vast quantities of excess human fat that it finds on its journey through the suburbs.

And here I thought the only thing that fat was good for was making alien babies on Doctor Who.
eatsuburbanites.jpg
Suburb-Eating Robot (PDF) [Andrew Maynard Architects]
]]>
http://io9.com/376647/suburb+eating-robots-run-on-fat-reclaimed-from-liposuction http://io9.com/376647/suburb+eating-robots-run-on-fat-reclaimed-from-liposuction Mon, 07 Apr 2008 07:00:00 PDT Annalee Newitz http://io9.com/index.php?op=postcommentfeed&postId=376647&view=rss&microfeed=true
<![CDATA[Dogoid Robot with No Head Moves in an Eerily Lifelike Manner]]> It looks like something out of a James Thurber illustration, with its headless body and backward-dog legs, but the Big Dog is real and autonomous. This video, by the Big Dog development team at Boston Dynamics, shows just how lifelike the bot is. It can carry over 300 pounds, and its engines make an alien whining noise. It can also, apparently, recover its balance after being kicked by its owner, in a scene that that is disturbing on a number of levels.

Something about this video made me think of nice black metal, perhaps from Sweden. I'm not sure why. So I made a little snippet with a new soundtrack.

New Video of BDI's Robot [IEEE Automaton]

]]>
http://io9.com/368949/dogoid-robot-with-no-head-moves-in-an-eerily-lifelike-manner http://io9.com/368949/dogoid-robot-with-no-head-moves-in-an-eerily-lifelike-manner Mon, 17 Mar 2008 15:30:26 PDT Annalee Newitz http://io9.com/index.php?op=postcommentfeed&postId=368949&view=rss&microfeed=true