Robot brains that think, plan and predict are being developed by a team of researchers in the US in a bid to create artificial brains.
Using the biological process of evolution researchers led by Chris Adami, a computational biologist at Michigan State University, are planning on rivalling human intelligence with robot brains that can navigate mazes, catch falling objects and collectively figure out how to exit and re-enter a room.
“Previous attempts to design human-like intelligence have failed because we don’t understand how our own brains work,” Adami said. “But we know how evolution works and we can speed it up inside of a computer.”
The researchers use genetic algorithms, called Markov networks, to program a large number of robot brains to work together on a particular task, like finding the exit to a maze. The brains that are quick to find the exit – or close enough, will have a large number of artificial “offspring”.
Once the algorithm is run over thousands, and often hundreds of thousands of generations, Adami’s team downloads the surviving brains into robots ready to carry out the task in the real world.
Evolving robot brains that rather learn about the world by interacting with it and could develop at the same speed as people’s is the best path towards self-aware intelligence according to Adami. “When robots have to make models of other robots' brains, they are thinking about thinking,” he said. “We believe this is the onset of consciousness.”
Talking about people’s apprehensiveness when it comes to robots, Adami was quick to dismiss it by saying: “When our robots are ‘born’, they will have a brain that has the capacity to learn, but only has instincts. It will take a decade or two of exploration and training for these robots to achieve human-level intelligence, just as is the case with us.”
Adami’s research was presented during the APS March Meeting in San Antonio, Texas.