The robot that BREEDS: ‘Mother’ machine builds and tests ‘children’ models to make each generation better than the last

Roboticists built a 'mother' robot (right) image ww.spy-drones.com

  • Roboticists built a ‘mother’ robot that independently builds ‘child’ models
  • It tests each ‘child’ to see which ones perform best at certain tasks
  • This ‘mother’ then uses the results to inform the design of the next ‘child’
  • Results show that preferential traits are passed down from one generation to the next – similar to how natural selection works in animals
During five experiments, the 'mother' designed, built and tested generations of ten 'children'image www.spy-drones.com

It may sound like the stuff of terrifying dystopian science fiction, but researchers have created robots that breed and evolve without any human interference.

The roboticists built a ‘mother’ machine that independently builds its own ‘children’ and tests which ones perform best at certain tasks.

This ‘mother’ then uses the results to inform the design of the next ‘child’, so that preferential traits are passed down from one generation to the next – similar to how natural selection works in animals.

Scroll down for video 

Roboticists built a ‘mother’ robot (right) that independently builds ‘child’ models (left). The the ‘mother’ robot was programmed to simply build a robot capable of movement and from this, and without any further human intervention or computer simulation, the robot built ‘children’ made of between one and five plastic cubes (left)

The robots were built by engineers from the University of Cambridge, lead by Dr Fumiya Iida.

Dr Iida and his team designed the ‘mother’ robot and programmed it simply to build a robot capable of movement.

From this, and without any further human intervention or computer simulation, the robot built ‘children’ made of between one and five plastic cubes with a small motor inside.

In order for the 'mother' to determine which 'children' were the fittest, each child was testedm image www.spy-drones.com

In each of five separate experiments, the mother designed, built and tested generations of ten children, using the information gathered from one generation to inform the design of the next.

The results found that preferential traits were passed down through generations, so that the ‘fittest’ individuals in the last generation performed a set task twice as quickly as the fittest individuals in the first generation.

HOW THE ROBOTS EVOLVED

Dr Iida and his team designed the ‘mother’ robot and programmed it simply to build a robot capable of movement.

From this, and without any further human intervention or computer simulation, the robot built ‘children’ made of between one and five plastic cubes with a small motor inside.

In each of five separate experiments, the mother designed, built and tested generations of ten children, using the information gathered from one generation to inform the design of the next.

The results found that preferential traits were passed down through generations, so that the ‘fittest’ individuals in the last generation performed a set task twice as quickly as the fittest individuals in the first generation.

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.

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.

‘Natural selection is basically reproduction, assessment, reproduction, assessment and so on,’ said Dr Iida 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.’

For each robot child, there is a unique ‘genome’ made up of a combination of between one and five different genes, which contains all of the information about the ‘child’s’ shape, construction and motor commands.

As in nature, evolution in robots takes place through ‘mutation’, where components of one gene are modified or single genes are added or deleted, and ‘crossover’, where a new genome is formed by merging genes from two individuals.

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.

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.

The researchers found that design variations emerged and performance improved over time.

The fastest individuals in the last generation moved at an average speed that was more than twice the average speed of the fastest individuals in the first generation.

This increase in performance was not only due to the fine-tuning of design parameters, but also because the mother was able to invent new shapes and gait patterns for the children over time, including some designs that a human designer would not have been able to build.

‘One of the big questions in biology is how intelligence came about – we’re using robotics to explore this mystery,’ said Iida.

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. The most successful individuals in each generation remained unchanged while mutation and crossover were introduced in the less successful children

ooo

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.’

In nature, organisms are able to adapt their physical characteristics to their environment over time.

The behaviour of the robots ('child pictured), is similar to that seen in nature. Organisms adapt characteristics to their environment over time

The behaviour of the robots (‘child pictured), is similar to that seen in nature. Organisms adapt 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.

But machines are not adaptable in the same way.

They are essentially stuck in one shape for their entire ‘lives’, and it’s uncertain whether changing their shape would make them more adaptable to changing environments.

Evolutionary robotics is a growing field which allows for the creation of autonomous robots without human intervention.

Most work in this field is done using computer simulation.

Although computer simulations allow researchers to test thousands or even millions of possible solutions, this often results in a ‘reality gap’ – a mismatch between simulated and real-world behaviour.

Iida’s research looks at how robotics can be improved by taking inspiration from nature, whether that’s learning about intelligence, or finding ways to improve robotic locomotion.

A robot requires between ten and 100 times more energy than an animal to do the same thing. Iida’s lab is filled with a wide array of hopping robots, which may take their inspiration from grasshoppers, humans or even dinosaurs.

‘It’s still a long way to go before we’ll have robots that look, act and think like us,’ said 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.’

The results are reported in the journal Plos One.

 ooo
Henry Sapiecha

Leave a reply

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>