However, the real story is told to us by Science Magazine. So what is all this fuss about robots? Well, this study tried to bridge between self- organization and evolution. Self-organization is the science that studies how collective phenomena (bees swarming, bird flocking, slime mold forming patterns, etc …) are induced by simple elemental rules at the individual level. Evolution is … well … the study of evolution.
Models of evolution have been normally used to explain the emergence and combination of very simple traits, neglecting often self-organization that, when studied together with evolution, can explain significant fitness boosts that would not appear without it.
On the quest to explain how complex self-organizing phenomena evolve, a group of scientists has been able to replicate through realistic simulation the evolution of self-organized division of labour. They modeled a scenario very similar to the one where atta leaf-cutter ants are involved. They used a special type of evolutionary computation approach called grammatical evolution to find the individual-level rules that yield to this type of organization.
Although pointing at the similarities between their model and biological systems, the same scientists also admitted that there are many key differences, first being evolutionary computation being only inspired by natural evolution and not modeling exactly that.
Future work aims at using more realistic evolutionary processes, while simplifying/sacryficing the robotics part.
The robot, ehm… the man behind all this is scientist (and friend) Dr. Eliseo Ferrante.
Before you go
If you enjoyed this post, you will love the newsletter of Data Science at Home. It’s my FREE digest of the best content in Artificial Intelligence, data science, predictive analytics and computer science. Subscribe!