Virtual, squishy creatures evolve to run using evolutionary algorithms

Apr 23, 2013

A research team led by Cornell University's Creative Machines Lab has created a computer algorithm that can be used to witness virtual creatures evolving their squishy, muscle-like features in order to teach themselves to walk.

The team incorporated concepts from and how nature builds complex animals – from to jaguars. The result is an array of bizarre, simulated robots that evolve a diverse series of gaits and gallops.

The paper describing these soft-bodied robots will appear in Proceedings of the Genetic and Evolutionary Computation Conference and is titled "Unshackling Evolution: Evolving Soft Robots With Multiple Materials and a Powerful Generative Encoding."

Lead author Nick Cheney, a Cornell graduate student, will present the paper at the July conference in Amsterdam. Robert MacCurdy, also a graduate student at Cornell, contributed to the work, as well as Jeff Clune, assistant professor at the University of Wyoming and former visiting scientist at Cornell.

This video is not supported by your browser at this time.

The video shows evolution in action: A creature evolves into a galloping, soft over 1,000 generations. While 1,000 generations is relatively short by natural evolution standards, it is enough to demonstrate the power of evolution to create counterintuitive designs, according to the researchers.

In the paper, they describe how they challenged human engineers to design robots made of these soft and hard materials. The human efforts paled in comparison to the designs resulting from evolution.

Explore further: New algorithm identifies data subsets that will yield the most reliable predictions

More information: Download the paper: jeffclune.com/publications/2013_Softbots_GECCO.pdf

More than a decade ago, Cornell professor Hod Lipson led a project called Golem that similarly evolved robots, and he later built them with a 3-D printer. That video and others from the Creative Machines Lab can be viewed at creativemachines.cornell.edu/videos .

add to favorites email to friend print save as pdf

Related Stories

Robots show the evolution of altruism

May 23, 2011

Scientists in Switzerland have pieced together the puzzle on the evolution of unselfish behaviour. They simulated genetic evolution over hundreds of generations by using simple robots, providing evidence of ...

Recommended for you

Designing exascale computers

Jul 23, 2014

"Imagine a heart surgeon operating to repair a blocked coronary artery. Someday soon, the surgeon might run a detailed computer simulation of blood flowing through the patient's arteries, showing how millions ...

User comments : 3

Adjust slider to filter visible comments by rank

Display comments: newest first

LariAnn
1 / 5 (1) Apr 23, 2013
A research team led by Cornell University's Creative Machines Lab has created a computer algorithm that can be used to witness virtual creatures evolving their squishy, muscle-like features in order to teach themselves to walk.


So this is really an example of directed or intelligence-driven evolution in action, as I'm sure the research team is intelligent and they did create the computer algorithm as indicated above - said algorithm didn't arise from random combinations of random bits of computer code with no help from the researchers.
lehman
5 / 5 (1) Apr 30, 2013
A research team led by Cornell University's Creative Machines Lab has created a computer algorithm that can be used to witness virtual creatures evolving their squishy, muscle-like features in order to teach themselves to walk.


So this is really an example of directed or intelligence-driven evolution in action, as I'm sure the research team is intelligent and they did create the computer algorithm as indicated above - said algorithm didn't arise from random combinations of random bits of computer code with no help from the researchers.


Yeah... this is an article about computational evolution...
antialias_physorg
5 / 5 (2) Apr 30, 2013
So this is really an example of directed or intelligence-driven evolution in action, as I'm sure the research team is intelligent and they did create the computer algorithm as indicated above

Well, no. You are not 'sure' because you didn't read the paper (it's linked at the bottom of the article for chrissakes. At least make an effort before posting, OK?)

The extension central to this paper - as opposed to previous papers - is that they allowed for multiple types of material during the evolution cycles: 2 types of supporting tissue: rigid (close to bone), squishy (soft tissue); and 2 types of 'muscles' (spontaneously contracting cells)

Selection was according to movement speed (which pretty much mirrors how selection works in real life for a lot of animals. Be slow and you get eaten)