Intelligent automated systems for processing and analysis of vehicle diagnostic data

Abstract

This article presents an approach to automating the process of vehicle technical condition diagnostics using intelligent automaton-based systems. The primary focus is on the application of automaton models, which enhance the accuracy of diagnostic parameter analysis in vehicles and reduce the likelihood of erroneous conclusions. The proposed system is based on spectral analysis methods, specifically orthogonal Fourier polynomials, which significantly reduce the root-mean-square error of diagnostic calculations. The study covers vehicle diagnostics via OBD-2, methods for retrieving and processing Diagnostic Trouble Codes (DTC), and the use of automaton models to assess the technical condition of vehicles. The main challenges in interpreting OBD data are discussed, and an approach combining automaton algorithms with spectral analysis methods is proposed to improve diagnostic accuracy. The proposed system has adaptive capabilities based on historical data, allowing it to autonomously adjust its diagnostic parameters depending on accumulated experience. The learning process is based on trend analysis, anomaly identification, and parameter adjustments in the automaton model to enhance diagnostic accuracy. Numerical modeling conducted as part of the study confirmed the effectiveness of the applied methodology. The developed intelligent automaton-based vehicle diagnostic system demonstrates high potential in the field of automated technical control. The use of automaton models improves the efficiency of diagnostic parameter analysis, reduces the probability of false conclusions, and enhances vehicle reliability. Future research prospects include expanding the application of probabilistic methods for handling large-scale diagnostic datasets.