This article discusses the current trends and prospects of robotizing technological processes in rail transport. Key technologies such as artificial intelligence (AI), computer vision, and predictive analytics are analyzed for their ability to automate maintenance, diagnostics, construction, and safety processes. Particular attention is paid to the use of robotic systems in various areas of rail transportation, including passenger services and logistics. The limitations associated with introducing robots are examined, and solutions are proposed. This article is based on an analysis of international experience and current data from 2024 to 2025.

The article presents a review of hybrid algorithms applied to the control of technological processes in railway transport. Special attention is given to the integration of artificial intelligence, optimization, machine learning, and physical modeling methods to improve the efficiency, reliability, and safety of railway systems. The paper discusses key types of hybrid solutions, including neuro-fuzzy systems, evolutionary algorithms with learning, digital twins, simulation–optimization approaches, and combined learning strategies. Examples of successful applications of hybrid models are provided for tasks such as dispatching, condition forecasting, schedule optimization, diagnostics, and energy efficiency. A comparative analysis of the accuracy, performance, and robustness of different approaches is presented. The advantages of hybridization over single-method techniques are highlighted, along with current challenges and future development directions.

The article provides a thorough analysis of the latest developments in China’s high-speed rail (HSR) sector, drawing from materials presented at the 12th World Congress on High-Speed Rail in Beijing in 2025. The article examines key technological achievements, such as the development of the innovative CR450 series of trains capable of reaching speeds of up to 450 km/h, the implementation of intelligent control systems (CTCS), and the shift towards next-generation communications technology (5G-R). Particular attention is given to the industry’s digital transformation, including the use of artificial intelligence for diagnostics and forecasting, creating digital twins of infrastructure, and large-scale robotization of maintenance processes. The article analyzes the role of the Chinese Academy of Railway Sciences (CARS) as a system-forming element of the national innovation ecosystem that provides a full development cycle from fundamental research to industrial implementation. The importance of a state strategy that combines large-scale investment in research and development (R&D), the development of a testing base, and active patent protection of technologies is emphasized. This article is useful for specialists in transport engineering, railway automation, and digital technologies, as well as for transport industry management representatives interested in advanced international experience.

This paper explores the prospects for adopting autonomous operation of freight trains, focusing on key challenges and advancements in this area. Although particular success stories exist, widespread testing is currently hindered by the lack of unified standards for evaluating system performance quality and safety certification processes. Major issues involve ensuring reliable communication between automated and conventional infrastructure components, establishing standardized interaction protocols, and refining mathematical models of train dynamics. Illustrative case studies cover innovations like parallel container handling, unmanned yard equipment, and freight transport automation. Efforts emphasize resolving critical safety and functional aspects of autonomous technologies to extend their applicability in transportation.

The article examines the reception of a freight train at a Digital Railway Station (DRS), a project being implemented on the South Urals Railway since 2018. The project entails developing a modular structure facilitating real-time interaction between various subsystems. Among the components of the DRS is an Automatic Vehicle Inspection System (AVIS), which builds a digital model of the train and decides its route based on inspection findings. When a train arrives at the station, a sequential chain of technological operations is carried out. The choice of each subsequent operation depends on the results of prior steps. Therefore, when designing modern railway station automation systems, sequential finite state machines (SFSM), particularly Mealy and Moore models, can be effectively applied. Utilizing automaton theory in synthesizing microprocessor-based railway station automation systems has enabled identification of bottlenecks while ensuring system safety and structural optimization. Consequently, to investigate the DRS model, it is recommended to formalize its description using finite state machine notation aiming to optimize interactions between DRS modules and ultimately improve station efficiency. The input alphabet may encompass events (signals) from diagnostic and control systems, while internal states reflect stages of the operational process. The output alphabet represents potential decisions regarding station management, realized by transmitting information to respective hardware and software complexes and DRS modules.

The article discusses the problem of lack of frequency resource in the VHF and UHF bands for trunked radio networks used in critical industries such as civil aviation and rail transport. A method for optimizing calls in DMR Tier III networks based on dynamic load distribution between base stations is proposed. The developed software includes three modules: a data converter, an analytical module, and a visualization module. Mathematical models of information processing are presented, as well as the results of testing the system, demonstrating an increase in DMR Tier III network bandwidth.

This paper presents the OMLS-Benchmark (Open Multi-Level Skills Benchmark) assessment system, a two–stage framework for the comprehensive assessment of large language models in software engineering tasks. The aim of the proposed approach is to overcome the limitations of existing techniques that either measure narrow subtasks or apply a single level of complexity and do not capture the dynamics of engineering skill and the interactive nature of diagnostic reasoning. The described system covers nine domains (Back-end, Front-end, Mobile, DevOps, Data Analysis, Machine Learning, Big Data, IoT, embedded systems) and five levels of complexity, which allows you to stratify quality by domains and levels. A two–stage procedure is proposed: Stage I standardized tasks with multiple choice and a strict response format, Stage II scenario tasks with step-by-step checklists and an independent judge model. The Tier Accuracy and Domain Accuracy metrics have been formalized, the OPS integral indicator has been introduced; the variables of formula (1) have been disclosed. Artifacts are published for reproducibility: JSON schemas of tasks, Russianlanguage templates of projects and an evaluation script. eval_mc.py with a description of the input/output parameters. Experiments show: heterogeneity of quality between domains; decreased results when switching from tests with fixed options to scenario tasks; detailed diagnostic reports on outstanding checklist items. The OMLS-Bench can serve as a practical tool for comparing LLMs in engineering tasks and as a basis for purposefully fine-tuning models to specific areas. The initial large-scale assessment of ten modern large language models revealed a clear stratification of results by complexity levels and by area: larger-scale models demonstrated high accuracy in widely represented web-oriented areas, while specialized areas (mobile development, embedded systems) showed significantly worse performance. These observations highlight the importance of both the size of the model and the variety of subject data in training. OMLS-Bench provides a reproducible and extensible evaluation tool that can serve as a basis for the development of more reliable and domain-specific assistant engineer models. In the future, it is planned to develop the interactive phase, increase the realism of scenarios and finalize control checklists to bring testing closer to professional practice.