Meet The Robot That Keeps Moving With Broken Legs
A new breakthrough has determined a fast method of helping robots move with damaged limbs: Giving them a "childhood" training period.
When we think about horrifying <a href=http://www.escapistmagazine.com/news/view/139454-Second-Trailer-For-Neil-Blomkamps-Chappie>robot apocalypses, we're almost always thinking about movies like <a href=http://www.escapistmagazine.com/news/view/140444-Terminator-Genisys-Trailer-Has-John-Connor-Spoiler>Terminator, where androids keep coming for you no matter how many limbs you blow off. Apparently some young roboticists watched that movie as children and decided decided such motion could go well beyond science fiction. That's why, according to a new report published in Nature, we can watch a six-legged robot compensate for the effects of broken limbs and malfunctioning motors within minutes. The technology is highly effective at letting robots determine the best way of reaching objectives - but rather than wipe out the human resistance movement, they'll be used to support rescue missions in disaster zones.
"Robots will eventually provide tremendous benefits to society, especially if they can complete tasks too dangerous for humans to perform," the University of Wyoming's Jeff Clune explained. "But robots won't be effective in those situations if they can't adapt and continue on after being damaged."
The breakthrough isn't based solely on the robots themselves - we've had robots capable of advanced movement for some time. What's key is a new algorithim that lets the robot figure out different gaits and motions much faster. Normally when a particular approach stops being effective, the robot tests various ways of moving until it finds one works. The process can take hours, so Clune and his team found a more effective method: Giving the robot a simulated "childhood".
"It plays for a while," Clume continued. "It learns lots of ways to move its body, just like a child saying, 'Oh, I can walk on my tippy toes, I can hop on one foot.'" Despite being the most adorable way to imagine robot movement this side of Big Hero 6, the process creates an intuitive movement sense that the robot relies on in the future. It also removes the need to program different movement scenarios into the robot, since it's capable of learning them itself.
What's more, the robot has already proved it can work under pressure. When the study's lead authors were asked to present the robot to VIPs, they had no idea custodial staff had waxed the floor to preparation for the demonstration. "It made the place look spic and span, but obviously the robot could no longer walk," Clume said. "But luckily our robot can deal with unforeseen situations. It adapted from sliding around the room to walking steadily in minutes."
The technology can even go beyond robots, as the team showed using a mechanical arm that keeps working with malfunctioning motors. The next step is to extend the algorithim to more complex technology used outside of the lab, such as search-and-rescue operations.
Source: Washington Post
Permalink
A new breakthrough has determined a fast method of helping robots move with damaged limbs: Giving them a "childhood" training period.
When we think about horrifying <a href=http://www.escapistmagazine.com/news/view/139454-Second-Trailer-For-Neil-Blomkamps-Chappie>robot apocalypses, we're almost always thinking about movies like <a href=http://www.escapistmagazine.com/news/view/140444-Terminator-Genisys-Trailer-Has-John-Connor-Spoiler>Terminator, where androids keep coming for you no matter how many limbs you blow off. Apparently some young roboticists watched that movie as children and decided decided such motion could go well beyond science fiction. That's why, according to a new report published in Nature, we can watch a six-legged robot compensate for the effects of broken limbs and malfunctioning motors within minutes. The technology is highly effective at letting robots determine the best way of reaching objectives - but rather than wipe out the human resistance movement, they'll be used to support rescue missions in disaster zones.
"Robots will eventually provide tremendous benefits to society, especially if they can complete tasks too dangerous for humans to perform," the University of Wyoming's Jeff Clune explained. "But robots won't be effective in those situations if they can't adapt and continue on after being damaged."
The breakthrough isn't based solely on the robots themselves - we've had robots capable of advanced movement for some time. What's key is a new algorithim that lets the robot figure out different gaits and motions much faster. Normally when a particular approach stops being effective, the robot tests various ways of moving until it finds one works. The process can take hours, so Clune and his team found a more effective method: Giving the robot a simulated "childhood".
"It plays for a while," Clume continued. "It learns lots of ways to move its body, just like a child saying, 'Oh, I can walk on my tippy toes, I can hop on one foot.'" Despite being the most adorable way to imagine robot movement this side of Big Hero 6, the process creates an intuitive movement sense that the robot relies on in the future. It also removes the need to program different movement scenarios into the robot, since it's capable of learning them itself.
What's more, the robot has already proved it can work under pressure. When the study's lead authors were asked to present the robot to VIPs, they had no idea custodial staff had waxed the floor to preparation for the demonstration. "It made the place look spic and span, but obviously the robot could no longer walk," Clume said. "But luckily our robot can deal with unforeseen situations. It adapted from sliding around the room to walking steadily in minutes."
The technology can even go beyond robots, as the team showed using a mechanical arm that keeps working with malfunctioning motors. The next step is to extend the algorithim to more complex technology used outside of the lab, such as search-and-rescue operations.
Source: Washington Post
Permalink
