In today’s world, where warfare and emergency situations unfold in complex, dynamic, and hazardous conditions, the development of autonomous systems capable of operating effectively without constant human control has become especially relevant. One of such promising areas is the development of cognitive AI platforms for the autonomous navigation of unmanned aerial vehicle (UAV) swarms—groups of drones that operate in close interaction with each other under artificial intelligence control, similar to a single organism.
In Ukraine, this idea is being implemented by a group of prominent scientists: Academician of the National Academy of Sciences of Ukraine (NASU) Mykhailo Zgurovsky, Corresponding Members of NASU Pavlo Kasyanov and Nataliia Pankratova, and their young students at the Educational and Scientific Complex “Institute for Applied Systems Analysis” of the Ministry of Education and Science of Ukraine and NASU. The creation of this unique system became possible thanks to the support of the National Research Foundation of Ukraine under grant No. 2025.06/0022. The developed platform is an example of deep integration of artificial intelligence, cutting-edge electronics, and strategic engineering thinking.
The cognitive AI platform represents a holistic system consisting of two complementary components. The ground center acts as the brain, where drones are trained, run simulations, and prepare for real-world missions. Each drone’s onboard system includes a set of intelligent modules that enable the aircraft to make decisions independently, navigate terrain, avoid obstacles, and coordinate with other drones — even in the absence of satellite signals or centralized communication. Thanks to this architecture, each drone functions as an autonomous agent with elements of self-learning. It can not only operate within the swarm but also distribute tasks, analyze sensory data, and adapt its behavior based on experience from previous missions.
The functionality of the cognitive platform has been tested in a number of realistic scenarios simulating combat or emergency conditions. These include reconnaissance of targeted areas, search and rescue operations after disasters, target engagement, communication support under electronic warfare conditions, and even disinformation missions against adversaries. One illustrative example is a simulation of a search and rescue operation after an earthquake: a swarm of 16 drones autonomously surveyed an area of 30 square kilometers, detected 27 thermal signatures, 21 of which turned out to be actual victims. The entire operation lasted less than an hour, and the swarm operated fully autonomously — without centralized control.
The uniqueness of this platform is defined by several key characteristics. First and foremost, it provides full autonomy — the drones can operate without GPS and constant contact with the operator. Secondly, the system’s cognitive nature is manifested in its ability to learn from experience by analyzing the outcomes of previous missions and improving behavioral models. Thirdly, the system’s architecture is scalable — it allows for the simultaneous coordination of both small teams and large swarms consisting of dozens of devices. Finally, the platform is highly flexible and resilient — it can adapt to environmental changes, respond to unpredictable conditions, and maintain system viability even in the case of loss or damage to individual drones.
Given its potential, the cognitive AI platform can be applied not only in military contexts. It can also be used effectively in peacetime — during disaster response, for environmental monitoring, medical supply delivery to hard-to-reach areas, protection of critical infrastructure, and more. The development of such a platform is a significant contribution to the advancement of next-generation autonomous intelligent systems. Ukrainian scientists are demonstrating their ability to create world-class innovative solutions that not only meet the contemporary challenges of security, defense, and humanitarian response but also open new horizons for technological progress.
Mariya Volynska