“Natural selection is basically reproduction, assessment, reproduction, assessment and so on. That’s essentially what this robot is doing – we can actually watch the improvement and diversification of the species.”
– Dr Fumiya Iida of Cambridge’s Department of Engineering
Showing an improvement over time in both design variance and performance, these Cambridge researchers have unleashed a new revolution in robotics – one in which autonomous robotics independently designs their successors by mimicking the biological process of natural selection. By testing her robot children on how far they’re capable of traveling, the mother robot learns from her previous mistakes and begins designing her next child with increased capabilities.
Bio-inspired robotics is a necessary evolution for the field if we’re ever hoping to acquire future robots that’ll not only pick up the task humans left off, but equally adapt quite efficiently to their surrounding terrain. The learning process for humans can be difficult, but is consistently in the realm of possibility. It’s an evolutionary trait that’s kept us alive for over 200,000 years. For robotics, up until now, this hasn’t been a trait well acquired. With their new evolutionary leap, however, expect the future of robots to not only outsmart us someday, but equally out-survive us as well. If our species wishes to survive the coming age of robotics, then we must be willing to merge with our new mechanical friends, thus creating a brand new species set to replace humanity!
Photo Credit: Michael Whelan