Swarm intelligence: Lessons from bees and fish for AI and next-gen robotics

Swarm Intelligence: Lessons From Bees and Fish for AI and Next‑Gen Robotics
A swarm of bees is far more intelligent than any single insect; now, AI and robotics are drawing lessons from these self-organising systems to design more efficient computing models.
Can a swarm of bees or a colony of ants be considered intelligent? To answer that, we first need to define what intelligence really means. Rather than an abstract quality, it may be better understood as the ability to adapt to one’s environment . Bees, for instance, communicate and share vital information to ensure the survival of the hive, despite lacking a complex language, cultural transmission or mastery of tools. They instead rely on a different form of intelligence: swarm intelligence , in which the whole becomes greater than the sum of its parts. Recent scientific research is now seeking to translate this efficiency into new approaches in robotics and artificial intelligence, as a striking example of biomimicry.
An international team of scientists from Penn State University in the United States has taken note of such interactions. Their proposal: microrobots capable of communicating through acoustic waves, forming swarms with collective intelligence that can restore their original formations after dispersing. The potential applications range from precision medicine to waste clean-up and the exploration of hostile environments .
Bats, whales and even bees have been making use of vibroacoustic signals for millions of years to communicate and navigate. Inspired by these natural systems, researchers in the US have created digital models of microrobots that emit and capture sounds, enabling hundreds of units to coordinate and move as a single entity—mimicking a flock of birds or a school of fish .
Each robot is designed to be extremely simple: a motor, a microphone, a loudspeaker and an electronic oscillator. Yet simplicity is deceiving; through sound-based communication, the robots synchronise their movements and adapt collectively to obstacles, dynamically reshaping themselves in response to the surroundings . The mechanism lies in each unit’s ability to adjust its oscillator to align with the swarm’s shared frequency, allowing it to move towards areas where the acoustic signal is strongest.
The most remarkable outcome comes from interaction: although each robot has no advanced intelligence, together they demonstrate emergent behaviours that allow complex tasks to be performed autonomously . This collective intelligence gives the swarm the ability to reorganise itself after disruptions—for example, deforming to pass through tight spaces and then reforming afterwards. The result: continuous operation without the need for external control.
Here are some possible applications of these swarming microrobots:
Cleaning up toxic spills or oil slicks in environments too dangerous for human access.
Acting as advanced, resilient sensors to detect threats, while maintaining function even if damaged.
Precision medicine , delivering drugs directly to target areas of the body by navigating thanks to their self-organising behaviour.
For now, this technology remains at the simulation stage. Researchers have tested the concept virtually with models of autonomous agents equipped with acoustic transceivers. Although these are not yet physical microrobots, the results suggest that collective intelligence and reconfiguration behaviour would carry over into real-world experiments.
“This breakthrough represents a significant step toward the development of smarter, more resilient microrobots based on low-complexity systems,” explains Igor Aronson, lead investigator at Penn State University. His team’s work, in collaboration with scientists at Ludwig Maximilian University in Munich, lays the foundations for a new generation of robots capable of autonomous adaptation in complex, unpredictable environments—able to meet challenges that once seemed out of reach.
While researchers at Penn State have focused on acoustic signalling, UK scientists have looked to bees’ visual processing to guide next-generation AI models. Recent work at the University of Sheffield shows that, despite their tiny brains, bees can learn and recognise complex visual patterns by actively combining flight movements with visual perception. What if AI could exploit a similar strategy?
The idea is that by integrating body–environment interaction, intelligent systems could master complex tasks without relying on massive computational power . Just as bees achieve efficiency in flight and vision, AI could adopt lightweight, agile models inspired by the deliberate simplicity of natural systems.
If you want to uncover further examples of biomimicry , check out our article on energy efficiency strategies inspired by animals and insects—where, once again, bees take centre stage.
Sources:
·        Physical Review Journals
·        ELife Sciences
David is a journalist specializing in innovation . From his early days as a mobile technology analyst to his latest role as Country Manager at Terraview , an AI-driven startup focused on viticulture, he has always been closely linked to innovation and emerging technologies.
He contributes to El Confidencial and cultural outlets such as Frontera D and El Estado Mental , driven by the belief that the human and the technological can—and should—go hand in hand.
I accept the  Information on data protection
Information on data protection
In compliance with Regulation (EU) 2016/679 on Data Protection and with other Data Protection regulations in force, you are hereby informed that your personal data shall be processed by Acciona, S.A. (hereinafter “ACCIONA”), whose identification data are as follows: Tax ID No. (NIF): A08001851; Address: Avenida de la Gran Vía de Hortaleza, 3, 28033 (Madrid); Tel. No.: +34 91 663 28 50; email: protecciondedatos@acciona.com. Your data shall be processed in order to send you information, through the subscription to our Newsletter through electronic means, regarding our latest technological innovation of ACCIONA´s initiative "I’MNOVATION #Hub", which aims to bring the general public main news, advances and curiosities of the innovation technology of all the companies that make up the ACCIONA Group.
The consent given by the data subject by indicating that they have read and accept this data protection information comprises the lawfulness of processing. The data provide is your email address. If you fails to provide the required data, the subscription Request cannot be satisfied. We may also process your satisfaction or preferences, if you voluntarily respond to surveys. Data shall be stored until the elimination is requested.
In order to carry out our purpose, we may give access to your data to service providers (such as technology service providers) who assist us in fulfilling this purpose. Some of its service providers, may be located outside of the European Economic Area in territories that do not offer a level of data protection that is comparable to that of the European Union. In such cases, we transfer User data with appropriate safeguards and always ensuring the security of the same.
The data subject can exercise their rights of portability and to access, correct, eliminate, restrict and oppose the processing of their data to Acciona, S.A. by writing the Department of Data Protection located at Avenida de Europa, 18, 28108, Parque Empresarial La Moraleja, in Alcobendas, Madrid or by sending an email to the following address:  protecciondedatos@acciona.com . If we consider it necessary in order to be able to identify you, we may ask you for a copy of an identity document. Furthermore, at any time the data subject may withdraw the granted consent by contacting the aforementioned address and file a claim to the Supervisory Authority (Agencia Española de Protección de Datos  www.aepd.es ). You can also unsubscribe from the Newsletter communication.
For any further information you can visit t…

Text scraped from the landing page for research purposes. © respective owners. This text is sourced from the advertiser's public landing page; for removal, contact dmca@luba.media.