When humans fall , they tend to have a sense of ego - preservation . They mystify out an arm , or a wooden leg — anything to obviate hitting the ground nose - first . Robots ? Not so much .
But researcher at the Georgia Institute of Technology are teaching automaton to fall graciously , saving time and money for roboticists whose prize research undertaking might infract its cervix — or its motor — assay to perform a undertaking that might seem simple to a human , like walk over uneven terrain .
Even the fancy , voguish robots come down sometimes . Just catch this gleeful compilation of cutting - edge ‘ bots break apart into the priming coat at the DARPA Robotics Challenge :
Thechallengewas originally found in 2011 , just after the nuclear disaster at Fukushima , as a manner to encourage development of the form of robots that could finally replace humans exercise in extremely dangerous areas — like repairing a downed atomic reactor . In the places where these kinds of robots would be most useful , cleaning up and offer human-centered attention after natural catastrophe , for instance , they ’re also most likely to take a tumble over something unexpected . Far from the research lab , they need to be able-bodied to get back up again .
The Georgia Tech algorithm allow a automaton to calculate how to arrive at the ground with less military force , so it does n’t burst itself . An accelerometer in the robot ’s question and a movement - gaining control camera are the queasy system , in essence , kick in it something akin to a human ’s reflexes . Instead of fall however gravity takes it , the robot attempts to make more than one contact item with the ground , dissipating some of the energy of the fall .
So far , the algorithm has only been tested on one robot , and in simulations with another , but given how many of the DARPA contest participant let graveness get the best of them , there ’s plenitude of test subject to work with in the future .
[ h / t : MIT Technology Review ]
Banner image screenshot viaYouTube
