Unmanned Aerial Vehicles

This paper investigates the impact of architectural scaling in YOLO-family neural object detectors on object detection performance in unmanned aerial vehicle (UAV) systems under CPU-only inference conditions without hardware acceleration. Standard nano and small model configurations are analyzed, along with an intermediate model obtained through controlled width scaling of the network. Experimental evaluation is conducted on an embedded Raspberry Pi 5 platform under fixed hardware and software conditions using ONNX Runtime, ensuring fair comparability of the models.

This article is licensed under the Creative Commons Attribution-NonCommercial 4.0 International License (CC BY-NC 4.0)

This paper addresses the current problem of flight planning in swarm systems of mobile sensor platforms with an integrated system for detecting camouflaged objects in complex observation environments. Considering the rapid increase in data volumes received from distributed sensor agents, as well as high requirements for accuracy and operational decision-making, a comprehensive approach to dynamic acquisition, processing, and recognition of three-dimensional images of structural objects is proposed.

In swarm missions of unmanned aerial vehicles (UAVs), one of the critical requirements for the control link is to ensure simultaneously long transmission range, low latency, and robustness to packet loss. Long-range radio communication protocols (LRS), such as ExpressLRS (ELRS) and TBS Crossfire, achieve a high level of interference immunity through the use of modern modulation techniques (LoRa/FSK) and adaptive data rate control.

The main prospects for the group use of unmanned aerial vehicles (UAVs) were analyzed, the problems were identified, and the stages of its solution were proposed. The basic algorithms are analyzed, which are designed to achieve joint and coordinated actions of all elements of the system of integrated control of a group of UAVs, and are used in most approaches to ensure stable interaction between system components.

The article considers an electronic system for detecting unmanned aerial vehicles, which includes a radar system, an optoelectronic system, a thermal imaging system, and an acoustic system. In information technology, the development of a software discrete-continuous stochastic model of the operational behavior of an electronic complex is an important stage in the creation of a structural-automatic model.

Aim of this research is to determine the area of land flooding according to a reasonably chosen hydrological model for a complex section of the Dniester River at the place of transition from the foothill to the plain with complex meandering and significant shifts of the river. Methods. The method of investigation of flooded areas as a result of water rise to a certain level has been processed.

The  use  of  state-of-the-art  technologies  for  the  standardization  of  work  related  to  environmental monitoring,  technological  processes,  urban  and  rural  infrastructures  requires  an  entire  system  of metrological  supervision. The drone is an unmanned aerial vehicle. The purpose of such systems of autonomous action, intended for flight and their performance of some operations can be potentially dangerous to humans. With the use of appropriate electronic equipment one can determine the level of air pollution and conduct the research.

The article is devoted to solution of the optimal design tasks of Unmanned Aerial Vehicles at different stages. The tasks for all UAV’s design stages are considered and CAx tools for optimal design are proposed based on empirical experience.