What is Swarm Intelligence?

schwarmintelligenz-vögel-comp

The group solves problems that are unsolvable as an individual. But how is that possible? The answer is swarm intelligence.

Swarm intelligence is an emergent phenomenon. It describes the collective behavior of decentralized, self-organized systems, natural or artificial. This event can be found in a wide range of areas, such as at birds, fish and ants, on the Internet and in humans.

A single stickleback does not manage to swim in shady regions but floats around in the light. His skin glitters in the light and he can be easily detected by enemies. But an entire swarm of sticklebacks manages to always follow the shady regions. If a swarm of sardines is attacked by dolphins, the fish form a sphere, the geometric body with the smallest possible surface. Moreover, they almost always get to escape the dolphins and immediately congregate again after a separation.

Similar phenomena can also be found in our everyday life. There are many internet sites which people speculate about the stock market on. A single statement remains useless but a large amount of speculation happens to be meaningful. Likewise the weight of a bull can be determined very precisely simply by calculating the average of the estimates of many humans.

And ants are just another example of swarm intelligence. They live in million-sized colonies and still have an admirable organization without any leader. A colony finds solutions to problems that would be unthinkable for individual ants. The colony uses the shortest way to the best source of food, distributes tasks to workers, defends the territory. An individual ant is helpless but a colony reacts quickly and efficiently to their environment. How is that possible?

The answers to this question are highly interesting – but also largely unknown. It is difficult to understand how communication between so many individuals works that smoothly. Locusts have shown that physical contact with other species is crucial for the mass to converge. The solidarity is controlled by the distribution of special scent messages. Ants also use scents to mark the shortest routes to food sources. And fish don’t use any unusually good memory skills to swim in the swarm either. Herrings and codlings follow two simple patterns: Follow the fish in front of you and keep the speed of the fish next to you.

In the early 1980s, the US researcher Brian Partridge discovered that a swarm basically acts like a sensor system in which each animal receives information and processes it independently of its peers. He discovered that the collective needs a threshold of 5% for the amplifying effect to begin. As soon as a group crosses the 5% threshold, the other animals follow as it is relatively likely for the collective that it will not fall victim to the error of an individual.

What applies to shoals of fish also applies to crowds of people. Pedestrians in a crowded pedestrian zone do not move wildly in confusion: lanes are formed. Each individual wants to make progress as quickly as possible without wasting unnecessary time on evasive maneuvers. In 2007, behavioral biologists from the University of Leeds, in cooperation with WDR, performed a human swarm experiment in which the participants had to stay in constant motion and an arm’s length away from their neighbors. After initial chaos, a rotating circle formed after some time. Such circular patterns are also known from some fish species.

Thus it can be said that we we, as is often the case, can learn a lot from mother nature . Nowadays, we investigate swarm intelligence for traffic, the calculation of routes as well as for robotics. If we apply the behaviour of swarm intelligence to our current problems, we can achieve a lot. Theoretically, we should be able to avoid traffic jams just as pedestrians avoid traffic jams in pedestrian zones, right?

 

Sources: focus.de, wikipedia.org, nationalgeographic.de, spiegel.de, deutschlandfunkkultur.de, youtube.com, scholarpedia.org, planet-wissen.de, berliner-zeitung.de, spektrum.de, gabler.de, wikipedia.de

Share this post

Share on facebook
Share on google
Share on twitter
Share on whatsapp
Share on linkedin
Share on pinterest
Share on reddit
Share on print
Share on email

Leave a Reply

Close Menu