KNITU-KAI: Technology for organizing the interaction of unmanned aerial vehicles in the search for ground objects

Main article: What is the Internet of Things (IoT)

2025: Creation of technology for organizing the interaction of unmanned aerial vehicles in the search for ground objects

Researchers from Kazan National Research Technical University. A.N. Tupolev (KNITU-KAI) created an algorithm for the effective interaction of unmanned aerial vehicles (UAVs) when searching for ground objects. The system automatically groups drones, coordinates their flights and combines data from visible and infrared cameras to obtain high-quality images of the terrain. The university announced this on February 13, 2025.

𝓲

Фото: KNITU-KAI

The authors of the work are Candidate of Technical Sciences, Associate Professor of the Department of Aircraft Structures Nikita Levshonkov, graduate student of the same department Igor Nafikov and student Yana Laryukhina.

The developed algorithm has already shown effectiveness in solving various practical problems: from creating digital maps of the area and monitoring agricultural land to detecting forest fires and patrolling beaches and airports. A separate direction is the delivery of goods and medicines to hard-to-reach areas.

One of the key elements of the system is the image processing unit. The algorithm works with fragments of 256 × 256 pixels, combining data from visible and infrared cameras. For each section of the image, the processing parameters are individually adjusted, which allows you to take clear pictures even in difficult observation conditions.

The system automatically calculates the optimal time for the task using the formula T=t₁ + (S_int )/( W_T × n) + t₂, where t₁ is the flight time to the search zone, S_int is the area of ​ ​ this zone, W_T is the performance of one drone, n is the number of vehicles in the group, t₂ is the return time.

Special attention is paid to flight safety. The algorithm analyzes possible errors when entering a flight mission, checks drone trajectories for dangerous intersections, takes into account the presence of obstacles - from trees to buildings. A minimum safe flight altitude is set for each search area.

In emergency situations, the system is able to independently determine the safe zone for landing a group of drones. If one of the sets loses communication or fails, its tasks are automatically redistributed among the rest of the group.

Technical tests have confirmed the algorithm's ability to effectively combine data from different drones. During the experiments, the system successfully processed images both at the top of the frame, where the visible range channel is most informative, and at the bottom, where this channel loses efficiency.

Based on the results of the study, it is planned to create a new product for the commercial operation of unmanned aircraft.