Answers are grounded in this paper and cite their sources.
The study objective is to create a system for assessing the dynamics of rolling stock using machine vision technology. The paper presents a comprehensive study of the dynamics of a railway vehicle, with the focus on analyzing the lateral misalignment of the wheelset. The main attention is paid to mathematical modeling in “Universal Mechanism” software package and the use of machine vision technology for processing video data. The task the paper is devoted to is to determine the parameters of lateral misalignment using machine vision technology. The research methods include mathematical modeling (“Universal Mechanism”), spectral analysis, machine vision technology, image processing algorithms (Python), fast Fourier transform, numerical experiment. The novelty of the work is in the developed comprehensive approach to studying locomotive dynamics, combining mathematical modeling in “Universal Mechanism” software complex and the use of machine vision technology for video data analysis. The research results are the following: a wheelset model is developed that takes into account its geometric and dynamic characteristics. A simulation of lateral misalignment is carried out while moving along a straight section of the track at a speed of 85 km/h. Graphs of the wheelset movement and a spectral analysis are obtained, which reveal a dominant oscillation frequency of 1.3 Hz. The machine vision algorithm is successfully applied to the recording from the camera mounted on the locomotive bogie frame. The same oscillation frequency is recorded, which proves the adequacy of the model and the applicability of machine vision for diagnosing the dynamics of the railway vehicle under real conditions. Conclusions: the results of the paper demonstrate the potential of using machine vision as an alternative to traditional measuring systems, which can improve the safety and efficiency of railway transport operation.
No takes yet. Share an insight, caveat, or question.
Grebennikov et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: