In the night skies above Budapest, a dense formation of 100 autonomous drones moves with a fluidity that recalls flocks of birds or herds of wild animals. This aerial choreography is the product of more than a decade of research at Eötvös Loránd University, where scientists have drawn directly from the collective movement patterns found in nature to engineer a new generation of swarm-capable unmanned systems.
The drones operate without centralized control, relying instead on algorithms inspired by the behaviors of pigeons in flight, wild horses on the Great Hungarian Plain, and other species. Each unit makes real-time decisions on trajectory and collision avoidance, communicating only with its immediate neighbors. This design allows the swarm to adapt dynamically to changing conditions, maintaining cohesion even in dense aerial traffic.
“It’s very rare that you see some technology and you say it’s beautiful,” said Boldizsár Balázs, one of the researchers involved in the project. “In its theoretical core it resembles nature. That’s why the drones themselves don’t need to be pretty, but what they do is pretty because it resembles natural swarming behavior.”
The principle behind the system is decentralization. Once the drones receive their initial instructions, they require no further input from a ground station. “After the drones are told what to do, we can switch off the ground control station, we can burn it or whatever, throw it away,” explained Gábor Vásárhelyi, senior researcher at the university’s Department of Biological Physics. “The drones will be able to do what they have to do just by communicating to each other.”
This approach marks a departure from conventional drone operations, which often depend on pre-programmed flight paths or continuous remote oversight. By embedding decision-making capabilities within each drone, the Hungarian team has created a swarm that can navigate complex environments autonomously, with potential applications ranging from meteorology and land surveying to precision agriculture.
The researchers have demonstrated the feasibility of scaling their algorithm to support swarms of up to 5,000 drones. In agricultural trials, they are developing systems for precision spraying of crops, aiming to improve efficiency while reducing chemical waste. In logistics, autonomous swarms could coordinate deliveries across urban areas without overburdening centralized air traffic control systems.
However, the technology’s potential extends into domains that raise ethical and security concerns. Anna Konert and Tomasz Balcerzak of Lazarski University’s Faculty of Law and Administration have studied the risks associated with autonomous drones, particularly in military contexts. “When drones take over lethal actions, responsibility may shift from human operators to machines, leading to uncertainty about who should be held accountable if errors occur,” they wrote in an email. “This detachment could lower the psychological barriers to initiating force, potentially making war more frequent and brutal.” They caution that reducing the human cost of military engagements could “encourage more frequent military actions, leading to faster conflict escalation since fewer immediate human consequences would weigh against the decision to engage militarily.”
Such warnings underscore the dual-use nature of autonomous swarm technology. While the Hungarian team’s work is rooted in biological physics and inspired by the elegance of natural systems, the same principles that allow drones to navigate cooperatively could be applied to less benign purposes.
For now, the focus remains on civilian applications. The ability to coordinate thousands of drones without centralized oversight could transform aerial data collection, environmental monitoring, and infrastructure inspection. By studying the movement of animals, the researchers have unlocked a method for machines to share spatial awareness and adapt collectively, a capability that may prove critical as skies grow more crowded with unmanned aircraft.