A ‘mother’ robot which can build its own ‘children’, test which ones do best, then modify their design, has been developed by scientists at Cambridge University.
In an astonishing video released by the university, the robot was shown building its own ‘cube babies’ before watching as they take their first steps.
After monitoring the progress of its offspring, the mother-bot then selects which have performed the best and refines their design in the same way that evolution works in nature.
It is the first time that the process of natural selection has been successfullybuilt into a machine and marks an important step forward in the quest forartificial intelligence.
The robot has been programmed to build a cube-bot with a small motor inside which is capable of movement then refine the design based on observations of how efficiently it moves.
In five separate experiments the mother-bot was allowed to continue building her children until 10 generations had been created, each time using the ‘fittest’ bot to inform the design of the next child.
“Natural selection is basically reproduction, assessment, reproduction, assessment and so on,” said lead researcher Dr Fumiya Iida of Cambridge’s Department of Engineering, who worked in collaboration with researchers at ETH Zurich.
“That’s essentially what this robot is doing – we can actually watch the improvement and diversification of the species.”
The robot children were designed from five simple rules which governed its shape, construction and motor commands.
In order for the mother to determine which children were the fittest, each child was tested on how far it travelled from its starting position in a given amount of time. Each child took the robot about 10 minutes to design, build and test.
The most successful individuals in each generation remained unchanged in the next generation in order to preserve their abilities, while mutation and crossover were introduced in the less successful children.
Over time design variations emerged and performance improved, not just through fine-tuning, but because the mother-bot invented new shapes and movement patterns, including designs which a human could not have built.
“One of the big questions in biology is how intelligence came about – we’re using robotics to explore this mystery,” said Dr Iida.
“We think of robots as performing repetitive tasks, and they’re typically designed for mass production instead of mass customisation, but we want to see robots that are capable of innovation and creativity.
It’s still a long way to go before we’ll have robots that look, act and think like us,” said Dr Iida. “But what we do have are a lot of enabling technologies that will help us import some aspects of biology to the engineering world.”
In nature, organisms are able to adapt their physical characteristics to their environment over time. These adaptations allow biological organisms to survive in a wide variety of different environments – allowing animals to make the move from living in the water to living on land, for instance.
In contrast machines are stuck in one shape for their entire ‘lives’, and researchers have been unsure whether they could adapt in the same way as organic life. It is the first time that it has been shown outside of a computer simulation that robots can adapt.
The research was published in the journal PLOS One.