The article provides a critical review of a wide range of modern publications by domestic and foreign authors in the field of development and implementation of intelligent control systems and intelligent transportation systems (ITS) in railway transportation. It is noted, that the prospects for the introduction of intelligent control systems in railway transportation are promising, especially in terms of improving safety, reliability, and operational efficiency. These systems combine artificial intelligence (AI), machine learning, and predictive data analysis to optimize operations, manage traffic, and improve service protocols. At the same time, it is shown that there is a risk of being carried away by the "fashionable" trend of digitalization without a systems analysis of its necessity and possibilities in the application areas, i.e. without a conceptual design of the ITS operating environment. The article gives practical recommendations on how to reduce these risks based on the basic principles of end-to-end conceptual technology of digitization.

The paper proposes the concept of using Computer Vision data to monitor the occupancy of sorting tracks (KZSP) when straightening and diagnosing the longitudinal profile of the track. The principles of calculating the longitudinal profile of sorting tracks based on the data obtained as a result of the work of the KZSP are shown. The results are presented at a real implementation facility (Kinel station of the Kuibyshev Railway), as well as an analysis of the profile, taking into account the current alignment. In conclusion, a description of the effectiveness of the KZSP in optimizing labor costs in the automation of sorting processes and the ongoing repair of the sorting fleet tracks is presented.

The paper describes the concept of an «erroneous» class in the task of recognizing the numbers of railway moving units – wagons and locomotives. The problems of existing recognition algorithms are shown using the example of a system for monitoring and preparing information about the movements of wagons and locomotives at a station in real time (SKPI PVL RV). The main problem of recognizing wagons and locomotives has been identified, related to multiple inscriptions and visual defects not related to the inventory numbers of rolling stock. A solution to this problem is presented by applying the Yolov8s model.

In intelligent transport, when constructing 3D maps, there is often a need to improve the quality of global localization on a pre-built map, for example, for initial localization when turning on a robot without additional assistance in the form of manual setting of its position. To solve this problem, this paper proposes an original architecture of the hierarchical localization approach, characterized by a multimodal model of place recognition based on video camera and LiDAR data and a high-speed neural network algorithm for point cloud registration. A software implementation of the proposed architecture has been developed and open sources as a part of OpenPlaceRecognition library. Experiments have been conducted with open datasets of real mobile robots NCLT and ITLP-Campus, which demonstrated an operating time of less than 100 ms and a sufficient level of quality on an embedded Nvidia Jetson AGX Xavier device. The prospects of the approach for practical application are shown.

The article is devoted to the actual problem of integrating knowledge about transport infrastructure in a geographic information system. The authors substantiate the need for a deeper understanding and formalization of geospatial knowledge, their structure and interrelations to automate the analysis and assessment of the location of transport infrastructure. The article discusses the organization of geospatial knowledge in solving problems in which it is necessary to analyze the specifics of the terrain. Logical-semantic analysis allows you to describe the features of geospatial knowledge. The scientific and applied value of geospatial knowledge in this context lies in the possibility of optimizing various processes related to transport infrastructure. The approach proposed by the authors to the modeling of geospatial knowledge and its computer implementation includes not only formal methods of knowledge representation, but also methods of processing real geospatial data. The authors pay attention to the issues of visualization of the analysis results in order to ensure the ease of use of the developed systems for specialists of different levels of training.

The main aspects of the operation of a high-speed diagnostic train and its role in ensuring the safety of high-speed railways are considered. Examples of diagnostic trains in different countries are given. The technical devices on the high-speed diagnostic train "IRIS 320" are considered. The analysis of the expected impact of the high-speed diagnostic train on the efficiency of the current maintenance of the track on the VSZHM-1 Moscow-St. Petersburg is carried out. The calculation of the frequency of rail monitoring on the VSZHM-1 by non-destructive methods is considered. Recommendations for the minimum and maximum number of inspections of rails by non-destructive testing methods are given. The technological challenges that will have to be faced when implementing a high-speed diagnostic train on the VSZHM-1 are considered and the problems in case of abandonment of this technical unit are described.

Hydraulic drive is widely used in the construction of ground transportation and technological facilities. Remote diagnostic methods with automated assessment of the technical condition of hydraulic drive components and assemblies in real time and instant transmission of control results to the service engineer's computer allow continuous monitoring of a remote facility, reduce downtime due to technical reasons, predict and prevent impending failures. The purpose of the work is to explore the possibilities of creating a system for remote diagnosis of the technical condition of hydraulic drives of mobile machines based on domestic components. The authors used an analytical research method, established a list of parameters for the technical condition of the hydraulic drive, examined the typical structure of the remote diagnostics system, for which they selected models of domestic sensors, devices and data exchange protocols, and tools for developing user programs.