Robots learn to share, validating Hamilton's rule (w/ video)

Robots learn to share, validating Hamilton's rule
Small foraging robots used in the experiment supporting Hamil-ton's rule (Image: EPFL/ Alain Herzog)

Using simple robots to simulate genetic evolution over hundreds of generations, Swiss scientists provide quantitative proof of kin selection and shed light on one of the most enduring puzzles in biology: Why do most social animals, including humans, go out of their way to help each other? In next week's issue of the online, open access journal PLoS Biology, EPFL robotics professor Dario Floreano teams up with University of Lausanne biologist Laurent Keller to weigh in on the oft-debated question of the evolution of altruism genes.

Altruism, the sacrificing of individual gains for the greater good, appears at first glance to go against the notion of "survival of the fittest." But altruistic is found in nature and is passed on from one generation to the next. Worker ants, for example, are sterile and make the ultimate altruistic sacrifice by not transmitting their genes at all in order to insure the survival of the queen's . The sacrifice of the individual in order to insure the survival of a relative's is known as kin selection. In 1964, biologist W.D. Hamilton proposed a precise set of conditions under which may evolve, now known as Hamilton's rule of kin selection. Here's the gist: If an individual family member shares food with the rest of the family, it reduces his or her personal likelihood of survival but increases the chances of family members passing on their genes, many of which are common to the entire family. Hamilton's rule simply states that whether or not an organism shares its food with another depends on its genetic closeness (how many genes it shares) with the other organism.

Robots learning to share: an interview with Dario Floreano

Testing the evolution of altruism using quantitative studies in live organisms has been largely impossible because experiments need to span hundreds of generations and there are too many variables. However, Floreano's robots evolve rapidly using simulated gene and genome functions and allow scientists to measure the costs and benefits associated with the trait. Additionally, Hamilton's rule has long been a subject of much debate be-cause its equation seems too simple to be true. "This study mirrors Hamilton's rule re-markably well to explain when an altruistic gene is passed on from one generation to the next, and when it is not," says Keller.

Previous experiments by Floreano and Keller showed that foraging robots doing simple tasks, such as pushing seed-like objects across the floor to a destination, evolve over multiple generations. Those robots not able to push the seeds to the correct location are selected out and cannot pass on their code, while robots that perform comparatively better see their code reproduced, mutated, and recombined with that of other robots into the next generation - a minimal model of natural selection. The new study by EPFL and UNIL researchers adds a novel dimension: once a foraging pushes a seed to the proper destination, it can decide whether it wants to share it or not. Evolutionary experiments lasting 500 generations were repeated for several scenarios of altruistic interaction - how much is shared and to what cost for the individual - and of genetic relatedness in the population. The researchers created groups of relatedness that, in the robot world, would be the equivalent of complete clones, siblings, cousins and non-relatives. The groups that shared along the lines of Hamilton's rule foraged better and passed their code onto the next generation.

The quantitative results matched surprisingly well the predictions of Hamilton's rule even in the presence of multiple interactions. Hamilton's original theory takes a limited and isolated vision of gene interaction into account, whereas the genetic simulations run in the foraging robots integrate effects of one gene on multiple other genes with Hamilton's rule still holding true. The findings are already proving useful in swarm robotics. "We have been able to take this experiment and extract an algorithm that we can use to evolve cooperation in any type of robot," explains Floreano. "We are using this algo-rithm to improve the control system of our flying robots and we see that it allows them to effectively collaborate and fly in swarm formation more successfully."


Explore further

Robots shed more light on evolution (w/ Video)

More information: Waibel M, Floreano D, Keller L (2011) A Quantitative Test of Hamilton's Rule for the Evolution of Altruism. PLoS Biol 9(5): e1000615. doi:10.1371/journal.pbio.1000615
Citation: Robots learn to share, validating Hamilton's rule (w/ video) (2011, May 3) retrieved 24 August 2019 from https://phys.org/news/2011-05-robots-validating-hamilton-video.html
This document is subject to copyright. Apart from any fair dealing for the purpose of private study or research, no part may be reproduced without the written permission. The content is provided for information purposes only.
28 shares

Feedback to editors

User comments

May 04, 2011
Simpson's did it.

May 04, 2011
Perhaps this would be a useful tool for getting assembler nanobots to cooperate in the construction of other stuff, and/or clones of themselves.

May 04, 2011
A one in a billion chance of a dangerous mutation happening and another one in a billion chance for that dangerous mutation to survive/avoid initial detection before reproduction. In the scale of trillions of trillions, the impossible can become very likely.

Evolving nano-bots could easily be the end of us.

May 04, 2011
Evolving nano-bots could easily be the end of us.


Only if they can multiply without failsafes. We're not stupid enough to allow that. Make one scarce element necessary for their reproduction and you've eliminated all risk. They would never be selected against it either since the non-mutants would naturally be far more successful.

Input 1kg of Palladium : Receive 10,000,000,000 bots

May 04, 2011
Evolving nano-bots could easily be the end of us.


When one looks at all the development that's going on it's either that or we moving ourselves into machines.

May 04, 2011
Existing assembly and production technologies are beginning to plateau, partly because of the limitations of our human bodies, but also partly because of the limitations of technologies based on positional assembly.

Molecular nanomachines will open a whole new mode of possibilities for making breakthroughs in medical, engineering, space exploration, and labor markets.

We may one day be able to make macroscopic machines which have a significant portion of their entire mass assembled at the molecular level. i.e. machines with advanced electronic neural interfaces, better engines and generators built to molecular precision, and better food production assisted by nano and micro robotic pest control systems.

With our faster and faster super computers, it will be possible to produce algorithms to help find optimum design specifications for cell-sized nano-assembly robots with molecular tools to do work on the human body or other process.

May 04, 2011
Evolving nano-bots could easily be the end of us.


Only if they can multiply without failsafes. We're not stupid enough to allow that. Make one scarce element necessary for their reproduction and you've eliminated all risk. They would never be selected against it either since the non-mutants would naturally be far more successful.

Input 1kg of Palladium : Receive 10,000,000,000 bots


That very response reminds me of Jurassic Park.

May 04, 2011
Only if they can multiply without failsafes. We're not stupid enough to allow that. Make one scarce element necessary for their reproduction and you've eliminated all risk.
As jm eloquently stated directly above, 'this reminds me of Jurassic Park'.

They're evolving.... If you make a necessary component scarce, evolution will select for a reduction in the need, and eventually the elimination of that need.

May 04, 2011
They could be following an external code to replicate themselves; the code would never change and so the only mutations would occur from construction errors. Those mutations would not be passed down like genes because the construction information is separate and not subject to natural selection. You would transmit their replication code wirelessly to the swarm. This would only serve as a replication tool and not an evolutionary one. With sufficiently advanced supercomputers you could run genetic algorithm simulations to evolve new kinds of bots for different tasks, safely. Runaway replication could be contained by using scarce elements like I said before.

I'm just brainstorming possible solutions, I really have no idea how viable any of this is.

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more