I, Xenobot: DARPA-Sponsored Scientists Use Frog Stem Cells to Develop Self-Healing 'Living Machines'
A team of scientists from the University of Vermont and Tufts University have succeeded in creating the first "living machine" that has us wondering if our future robotic overlords might not be cyborgs at all, but hyper-intelligent amphibious mutants. The xenobots, as they are being called, were designed on a supercomputer at UVM that modeled simulations of certain passive and contractile cellular behaviors, then assigned the models simple tasks like moving or rotating. The models were realized by Tufts scientists who harvested stem cells from African clawed frogs and assembled them into tiny, living organisms based on the models that performed the tasks as "programmed." Their research, sponsored by DARPA, was published in the Proceedings of the National Academy Sciences of the United States of America (PNAS) journal.
If this sounds an awful lot like scientists "playing God," well, you're not far off. "These are novel living machines," said UVM robotics expert and research co-leader Joshua Bongard in a statement to UVM Today.
The key to the xenobots is in the models generated by UVM's Deep Green supercomputer cluster. Very simply, Deep Green uses an evolutionary algorithm to create thousands of models of tiny lifeforms based on simulated skin and cardiac cells and modeled around the achievement of a task, e.g., moving forward in one direction. As the computer discerns which models successfully accomplish the task, the scientists decide which ones they want to actually test. Those models were sent to Tufts, where scientists painstakingly joined single stem cells to build the models created by the algorithm. Sure enough, they worked just as the simulations predicted they would.
The xenobots are nothing like the Terminator or even your Roomba: there's no metal involved in them whatsoever. Unlike your Roomba or the cell phone you keep breaking the glass on, however, xenobots can heal themselves: a fact demonstrated on camera. "We sliced the robot almost in half and it stitches itself back up and keeps going," Bongard said. "And this is something you can't do with typical machines."
Addressing potential concerns about things going horribly wrong, Tufts researcher and project co-leader Michael Levin told UVM Today that such a fear "is not unreasonable. When we start to mess around with complex systems that we don't understand, we're going to get unintended consequences." Nevertheless, Levin argues that the project serves a larger purpose towards understanding such consequences. "I think it's an absolute necessity for society going forward to get a better handle on systems where the outcome is very complex," Levin continued. "A first step towards doing that is to explore: how do living systems decide what an overall behavior should be and how do we manipulate the pieces to get the behaviors we want?"
Although reframing that sentiment, Bongard seemed to agree with his colleague. "There's all of this innate creativity in life," he said. "We want to understand that more deeply-and how we can direct and push it toward new forms."
Whether or not that understanding is best accomplished by creating entirely new life forms shall remain to be seen.