This article addresses an approach to intelligent safety control of passengers on escalators. The aim is to improve the accuracy of detecting threatening situations on escalators in the subway to make decisions to prevent threats and eliminate the consequences. The novelty of the approach lies in the complex processing of information from three types of sources (video, audio, sensors) using machine learning methods and recurrent neural networks with controlled elements. The conditions and indicators of safety assurance efficiency are clarified. New methods and algorithms for managing the safety of passengers on escalators are proposed. The architecture of a promising safety software system is developed, and implementation of its components for cloud and fog computing environments is provided. Modeling results confirm the capabilities and advantages of the proposed technological solutions for enhancing the safety of escalator passengers, efficiency of control decision making, and system usability. Due to the proposed solutions, it has become possible to increase the speed of identifying situations 3.5 times and increase the accuracy of their determination by 26%. The efficiency of decision making has increased by almost 30%.
A model-based approach to large-scale distributed system architecting is suggested, which is based on the use of dynamic digital twins. This approach can be considered as an integration of known paradigms, such as digital twins, evolutionary architecture and agile architecture. It can also be considered as one of the possible realizations of the digital thread paradigm. As part of this approach, a three-level digital thread reference architecture is suggested, which includes the following levels: (i) digital thread support level; (ii) agile architecture support level; (iii) digital shadow support level. This approach has been used in the development of a number of real systems, and has shown its effectiveness in supporting system agility at the exploitation and modernization stages. The proposed approach is focused on building digital twin-based systems. This article may be interesting for specialists engaged in research and development in the domain of IoT- and IIoT-based information systems, primarily architects.
The paper focuses on distributed video processing for embedded computers using foggy environments. The possibilities of using foggy technologies for intelligent video processing surveillance are identified. A comparison of existing systems for video processing is provided. The place of embedded computers in the intelligent video surveillance systems and the ways of modernizing these systems are considered. The data from the devices is collected using Internet of Things technologies. A system for solving tasks of intelligent video surveillance has been developed and the effectiveness of the new system has been proven. As an example, the task of processing video received from the cameras used in the subways is considered. The proposed system can be applied not only for processing images in the subways but in many other domains where the Internet of Things technologies are used.
This article describes the application of machine learning methods to control the process of railway flaw detection. It is told about the new capabilities of computer technology, which make it possible to clarify the damage for the lineman. A new way of interaction of a specialist with diagnostic information using the Internet of Things is proposed. The description of architecture is given according to three standards at the design stage. An application for a smartphone has been developed. The effectiveness of using foggy computing environments and cloud technologies for recognizing damage to railway tracks using machine learning has been proven. Specific examples are given.