Ulyanovsk State Technical University (Russian: Ульяновский государственный технический университет, abbreviated as UlSTU) is a major technical higher education and research institution in Ulyanovsk. Founded in 1957 as Ulyanovsk Polytech University, the university is located in the Volga River region.Ulyanovsk State Technical University (UlSTU) is one of the leading science centers of the Volga region that trains specialists in various areas, including economy, humanities, technical sciences, etc.The university is well known in Russia for its research-and-development activities; graduates of UlSTU are employed by a large variety of companies. Today Ulyanovsk State Technical University has over 12,000 students in different educational programs at 12 faculties and 48 departments; more than 500 tutors, 42 Ph.D professors and 250 Ph.D associate professors; scientific and research works and papers awarded in Paris, Brussels etc.; and one of the biggest concert halls in the Middle Volga region.The University maintains firm contacts with more than 60 state, public and educational organizations of many countries: universities, research centers, foundations. UlSTU closely cooperates with such educational and scientific centers as Darmstadt University of Applied Sciences, Germany; Durham University, Great Britain; University of Göttingen, Germany.Sporting facilities, a calendar of social events and a reputation for teaching makes UlSTU a place to study. Students from Ukraine, Kazakhstan, Germany, China, Vietnam, Turkey, Ghana, Cameroon, and Nigeria successfully study at the University..
An innovative sixth algebraic order technique of the Adams-Bashforth-Moulton predictor–corrector–corrector type is introduced in this paper. The innovative method allows for the accurate integration of any set of functions [ 1, x, x^2, x^3, x^4, e^I v x] . The new method’s stability zones have been drawn. This new approach was used in quantum chemistry to solve the difficult Schrödinger-type coupled differential equations problem. Other related problems are also addressed with the novel method.
This study presents a novel low-order strategy for the Adams–Bashforth–Moulton predictor–corrector. The functions [ 1, e^I v x, x e^I v x] can be linearly merged utilizing this novel approach. The stability zones for the novel approach have been delineated. The challenging Schrödinger-type coupled differential equations problem was addressed in quantum chemistry using this novel method.
Hydrogen enrichment is a pivotal strategy to augment the combustion characteristics of biogas, addressing its inherent limitations of low calorific value and sluggish flame propagation. This study presents a rigorous numerical investigation into the synergistic effects of hydrogen blending ratios (10 %, 20 %, and 40 % by volume) and air excess ratios (alpha = 0.8, 0.99, 1.09, 1.19) on the thermochemical performance of a biogas-fueled system (65 % CH4 / 35 % CO2). The results demonstrate that hydrogen enrichment significantly enhances combustion efficiency and flame stability while facilitating thorough oxidation. A peak axial temperature of 1591 K was recorded for the 20 % H2 blend under stoichiometric conditions, representing the optimal thermal output. As the air excess ratio transitioned from sub-stoichiometric (0.80) to lean conditions (1.19), a controlled thermal reduction was observed, which effectively optimized flame stability and mitigated localized hotspots. Quantitatively, the study identifies alpha = 1.09 with 20 % hydrogen enrichment as the optimal operating configuration, yielding the best compromise between high thermal efficiency and minimal pollutant emissions. Under these conditions, CO emissions were significantly curtailed, while peak temperatures remained below the critical threshold for significant Thermal-NOx formation via the Zeldovich mechanism. These findings provide a high-fidelity physical benchmark for the design of sustainable, low-emission energy conversion systems, bridging the gap between traditional CFD analysis and future AI-driven combustion optimization.
Real-time analysis of unmanned aerial vehicle (UAV) imagery is an important task in precision agriculture. However, deploying deep learning models on embedded platforms is challenging due to limited computational resources and power constraints. This paper investigates the acceleration of agricultural image segmentation using YOLOv11 instance segmentation models, INT8 quantization, and Rockchip NPU deployment. A combined dataset consisting of real UAV images and synthetic samples was used for crop, weed, and soil segmentation. YOLOv11n-seg and YOLOv11s-seg models were trained, quantized, converted to the RKNN format, and deployed on an Orange Pi $5+$ platform based on the RK3588 processor. Experimental results show that INT8 quantization reduces segmentation accuracy by approximately $2-3$ percentage points while providing a $\mathbf{2. 6}-\mathbf{2. 7} \times$ reduction in inference latency. The obtained results demonstrate the feasibility of real-time agricultural video analytics on low-power embedded hardware and confirm the effectiveness of Rockchip-based edge AI platforms for UAV applications.
Разработана методика построения траектории беспилотного летательного аппарата, основанная на анализе видеопотока с расположенной на нём камеры без применения информации глобальных навигационных систем. Траектория строится по найденным параметрам аффинной модели взаимного пространственного рассогласования смежных видеокадров. Использованы методы нахождения особых точек Ши - Томаси, оптического потока Лукаса - Канаде и фильтр Калмана. Приведены результаты реализации методики на одноплатных компьютерах и её апробации в условиях ограниченных вычислительных ресурсов на реальных видеопоследовательностях. A method of unmanned aerial vehicle trajectory planning has been developed. This method is based on the analysis of a video stream from a camera mounted on the vehicle, without using information from global navigation systems. The trajectory is planned using the determined parameters of an affine model of mutual spatial misalignment of adjacent video frames. The method utilizes the Shi-Tomassi singularity detection method, the Lucas-Kanade optical flow method, and the Kalman filter. The results of the method implementation on single-board computers and testing it under limited computing resources on real video sequences are presented.
