Vol 24, No 3 (2025)
Robotics, automation and control systems
A Robust Control Algorithm for Single Input Single Output Dynamic Object Based on Table-Based Q-Method of Reinforcement Learning
Abstract
The article provides an overview in the field of dynamic object control systems based on reinforcement learning. Based on the analysis, it is concluded that the development of control methods based on reinforcement learning is relevant. The article proposes an intelligent algorithm for robust control of stable dynamic objects with one input and one output, based on the tabular Q-learning method of zero order. The algorithm stabilizes the output value of the control object with a given error if the parameters and external disturbances of the object are piecewise constant unknown quantities, and the state vector is measurable. The novelty of the proposed algorithm lies in a new incremental method of control formation, which allows, based on a set of three possible actions, to stabilize the control object. The proposed method of forming a set of control actions makes it possible to ensure the required accuracy of stabilizing the output of an object by changing the amplitude of the control increment. The proposed algorithm has high computational efficiency. After training, the control calculation is reduced to calculating indexes based on measurement results, reading data from memory based on calculated indexes, and finding the maximum value in a small vector. For a discrete description of the control object, the conditions of convergence of the learning algorithm and the limitation of the control error are investigated. The developed algorithm is demonstrated by the example of the synthesis of robust control of a DC motor with independent excitation. In the course of numerical simulation, the quality of a closed system is investigated when the parameters and the control action change. The analysis of the simulation results allows us to draw conclusions about the effectiveness of the synthesized algorithm. The article also provides the results of a real experiment that demonstrate the technical feasibility of the algorithm obtained. This issue is important, since the analysis of sources shows an almost complete lack of technical implementation of control systems for dynamic objects synthesized using reinforcement learning methods.
Informatics and Automation. 2025;24(3):717-744
717-744
Synthesis of Combined Quadrocopter Attitude and Altitude Control Based on Block Approach with Sigmoidal Feedbacks
Abstract
This paper considers the problem of controlling the attitude and altitude of a quadrocopter in the presence of uncertainties in the plant model. When solving this problem, it is especially important to consider the peculiarities of the plant: strong susceptibility to roll, pitch, and altitude oscillations due to the quadrocopter design and motor dynamics (with yaw being the least susceptible to oscillations due to motor dynamics compared with other controllable variables). To achieve high control quality in the presence of uncertainties, combined control is usually applied. It is constructed as a sum of two parts: a basic stabilizing part and a part compensating uncertainties with the help of a disturbance observer. Typically, both parts contain linear feedback. However, when the output variables of the plant track non-smooth reference signals, linear feedback can cause overshooting and increased oscillations. To prevent these problems, we propose a combined control law with smooth and bounded feedback in the form of the hyperbolic tangent. This feedback is used by both the controller and the disturbance observer. In this case, the control synthesis is based on the structural properties of the plant using the block approach. Its application provided invariance of the output variables with respect to not only matched but also unmatched uncertainties, and also allowed to construct a disturbance observer of the minimum possible order. In addition, to reduce the oscillations, a part with plant accelerations was introduced into the control law. To realize the proposed approach, it is sufficient to know the nominal values of some parameters of the plant and the permissible bounds of uncertainty variation. We present the results of experiments on a quadrocopter with an F450 frame and the results of a comparative analysis of the proposed approach with the one using linear control.
Informatics and Automation. 2025;24(3):745-790
745-790
Planning UAV Flight Trajectories for Monitoring Large Areas
Abstract
Modern agriculture covers vast areas, and effective monitoring of these territories plays a key role in precision farming. Wireless sensor networks are widely used to obtain real-time information on the condition of agricultural crops. However, manually collecting data from such sensors (deployed across a sensor network) is challenging. At the same time, unmanned aerial vehicles (UAVs) are increasingly used to provide automated and highly accurate data collection. This article addresses the problem of constructing an optimal UAV trajectory to efficiently collect data from distributed sensor nodes. The goal is to minimize the total route length while fully covering the sensing zones of all devices. Within the study, four route planning methods were developed and compared: the centroid-based method, the three-point method, the tangential method, and the optimal point selection method within the coverage radius boundary. Each method was implemented as a programmatic algorithm, including route construction, geometric optimization, and coverage evaluation. All methods were tested under the same conditions using a set of sensors distributed over a defined area. Evaluation criteria included total path length, number of maneuver points, and computation time, across coverage radii from 1 to 50 meters. The authors propose two approaches for trajectory optimization: a clustering-based centroid algorithm and an enhanced three-point algorithm based on the Lin-Kernighan heuristic. Experimental results showed that the proposed dual-algorithm method significantly outperforms previously studied route planning methods. Thus, this paper presents a comprehensive approach to UAV route planning for agricultural field monitoring, considering geometric, algorithmic, and computational factors. It also provides practical recommendations for selecting the most suitable method based on the spatial structure of the sensor network.
Informatics and Automation. 2025;24(3):791-827
791-827
A Comprehensive Analysis of Multi-Channel Mac and Clustering Protocols for Robust and Energy-Efficient Wireless Sensor Networks
Abstract
Wireless Sensor Networks have become indispensable in various applications, from environmental monitoring to health tracking. As they continue to evolve, security and energy efficiency remain paramount. This analysis paper compares contemporary techniques within two significant protocol categories: Multi-Channel Medium Access Control (MAC) protocols and Cluster-Based protocols. The evaluation focuses on various channel assignment strategies and clustering methods, including static and dynamic allocation of communication resources, adaptive methodologies, and hybrid approaches, alongside strategies for selecting and rotating cluster heads, and aggregating data efficiently. Through a comprehensive examination, we highlight the limitations and potential of each approach, proposing a hybrid framework that leverages the strengths of both protocol types to enhance security and energy efficiency in Wireless Sensor Networks. Our findings suggest that integrating dynamic resource allocation with energy-efficient clustering and adaptive strategies with rotational cluster head selection could lead to more robust and efficient deployments. This analysis serves as a foundational study for future research, aiming to develop advanced hybrid protocols that address the dynamic demands of WSNs, ensuring sustainable and efficient network operations.
Informatics and Automation. 2025;24(3):828-855
828-855
Optimizing Transportation Costs: Enhancing Logistics Efficiency and Resource Utilization in Dynamic Environments
Abstract
The increasing demand for goods transportation, driven by the expansion of global supply chains and rising customer expectations, underscores the critical need to optimize transportation costs to enhance logistics efficiency. In a rapidly evolving and competitive market, businesses face mounting challenges in managing complex transportation networks, minimizing operational costs, and meeting diverse customer requirements. To address these issues, this paper introduces a solution designed to reduce transportation expenses by optimizing the flow of goods and improving resource utilization. By leveraging advanced optimization techniques and data-driven strategies, the proposed solution identifies inefficiencies, streamlines decision-making, and enhances resource allocation. Initial results demonstrate that this approach not only significantly reduces operational costs but also strengthens the ability of businesses to respond quickly and effectively to fluctuating customer demands, ensuring both cost efficiency and customer satisfaction. However, as the logistics industry continues to grow and transaction volumes increase, transportation scenarios are expected to become more complex, and customer requirements more diverse. This evolving landscape demands further refinement and scalability of the proposed solution to address larger networks, more intricate logistics challenges, and a broader range of customer demands. Future research will prioritize the development of larger-scale models capable of incorporating more variables, improving computational efficiency, and delivering faster, more accurate decision-making to meet the increasing complexity of the logistics sector. Therefore, the proposed solution represents a significant advancement in optimizing transportation costs and improving logistics efficiency. Initial results indicate that this solution can cut down transportation costs by 19.02% to 29.65% and enhance computational efficiency in small- to medium-scale routing tasks (10–20 customers). Despite its potential, more research is required to justify scalability to larger datasets. Hence, our approach provides a solid foundation for logistics optimization, with clear prospects for expansion and adaptation in real-world contexts.
Informatics and Automation. 2025;24(3):856-883
856-883
Modified Heuristic Task Allocation Algorithms for Mobile Robot Teams under Uncertainty
Abstract
This study addresses the problem of task allocation among groups of mobile robots under conditions of parametric and stochastic uncertainty arising from sensor errors, environmental non-stationarity, and limited information about controlled objects. The primary objective is to adapt previously developed heuristic algorithms to real-world conditions, where sensor inaccuracies and incomplete knowledge of the environment are present. Three baseline approaches are considered: the ant colony algorithm, the Hopfield neural network, and the genetic algorithm. Each method is enhanced with specific modifications to account for input uncertainty: dynamic pheromone trail updates, adaptive adjustment of neuron weight coefficients, and interval-based estimation of environmental parameters. The paper presents a formal problem statement, mathematical models, and the design principles of the proposed task allocation algorithms. Numerical simulations were conducted to compare the performance of the modified algorithms against their baseline counterparts under varying levels of operational uncertainty. Results show that the proposed adaptive mechanisms improve task allocation efficiency by up to 20% compared to the original methods. Based on these findings, recommendations are formulated for selecting the optimal algorithm depending on specific operating conditions and control objectives. The study concludes that the proposed approaches are effective for the design of intelligent adaptive group control systems for mobile robots. Furthermore, these solutions can be extended to a broader class of problems, including dynamic resource reassignment and the organization of cooperative behavior among technical agents.
Informatics and Automation. 2025;24(3):884-913
884-913
Mathematical modeling and applied mathematics
An Approximate Assessment of Latency in a Computer System with Container Virtualization
Abstract
The key role in achieving high reliability, security, fault tolerance, and low latency of query service in distributed systems (including cloud computing) is played by the consolidation of data processing and storage resources in clusters, the efficiency of which increases with the use of virtual machine technologies and container virtualization. The complexity of building queuing models for container virtualization systems is caused by the fact that the intensity of query execution in each container is associated with the dynamic division of shared resources between active (performing functional tasks) containers and the costs of supporting all containers deployed in the VM, including inactive containers waiting for service requests to be sent to them. The reduction in service intensity in each container due to shared resource allocation depends on many factors that are difficult to investigate. For clusters with container virtualization, this article provides an approximate boundary estimate of the average request waiting time and the probability of timely service. When building an analytical model, each container is represented as a separate single-channel queuing system with an infinite queue and the simplest input stream. The key feature of the proposed virtual cluster model is the estimation of upper, lower, and average bounds for the potential service intensity reduction in containers, resulting from the allocation of a node's limited computing resources among them. This depends on the number of deployed containers and the dynamically varying count of active containers, which is influenced by the input stream intensity. The study demonstrates the existence of an optimal number of containers per node, minimizing the average request processing time or maximizing the probability of timely request execution. The proposed models can be applied to the structural and parametric optimization of clusters with pipelined virtualization, including in the case of scaling and reconfiguration adaptive to traffic changes by disconnecting or connecting some of the deployed containers depending on changes in the load in the system.
Informatics and Automation. 2025;24(3):914-950
914-950
Optimization of Integrated Energy System Resilience
Abstract
Currently, the development of approaches that enhance the resilience of integrated energy systems is a highly relevant research direction. Such approaches are based on the structural and parametric optimization of integrated energy systems. Typically, these approaches are closely tied to a specific spatio-temporal scope and a particular optimization method. The application of developed approaches at other scopes often leads to a significant increase in computation time and a possible reduction of solution accuracy. This problem is due to the complexity of energy system optimization models and the differences between them. To solve this problem, we have developed a methodology for selecting the most suitable methods for the design of system resilience at a given spatio-temporal scope. The proposed methodology is based on testing methods within a specialized testbed and a multi-criteria analysis of test results. The indicators for evaluating the methods include both summary metrics of resilience and efficiency parameters of computational resources. The benefits of the proposed methodology are illustrated for the resilient design with respect to national and local integrated energy systems. Several dozen methods from the well-known Parallel Global Multiobjective Optimizer library were efficiently tested in up to 10 hours. The analysis of the testing results was performed with different multi-criteria algorithms regarding the prioritization of the indicators.
Informatics and Automation. 2025;24(3):951-981
951-981
Comparative Study of Person Re-Identification Techniques Based on Deep Learning Models
Abstract
Person re-identification (Re-ID) is crucial in intelligent surveillance, requiring precise identification of individuals across multiple camera viewpoints. Traditional distance-based methods, such as Euclidean and Cosine, struggle with challenges like posture variations and occlusions, limiting their effectiveness. This study explores deep metric learning models, specifically Siamese and Triplet networks, to improve Re-ID performance. We evaluate these methods on the Market-1501 dataset using Cumulative Matching Characteristic (CMC) and Cumulative Distribution Function (CDF) curves. Our findings reveal that the Triplet network outperforms traditional approaches at higher ranks, achieving Rank-5 accuracy of 78.6% and Rank-10 accuracy of 93%, while its Rank-1 accuracy remains low (0.06%). In contrast, Euclidean and Cosine distances show poor Rank-1 performance (2% and 0.30%, respectively), highlighting their limitations. Additionally, incorporating VGG16 enhances feature extraction, improving recognition by capturing fine-grained spatial details. This comparative study highlights the effectiveness of deep metric learning and underscores its potential for real-world surveillance applications. However, the computational demands of deep networks present challenges for real-time deployment. Future research should focus on optimizing model efficiency, reducing computational costs, and extending evaluations to real-time scenarios.
Informatics and Automation. 2025;24(3):982-1001
982-1001