era to chart the robot's position and feed that information back into the robot's control program. But the re - searchers wanted to create a soft robot untethered from external aid. They developed a novel neural network architecture that both optimizes sensor placement and learns to efficiently complete tasks. First, they divided the robot's body into regions called " particles. " Each particle's rate of strain was provided as an input to the neural network. Through a process of trial and error, the network " learns " the most efficienthttp://info.hotims.com/79417-757 http://www.abpi.net/ntbpdfclicks/l.php?202109TBNAV