Comprehensive method for modeling the quality parameters of an autonomous vehicle

Modern autonomous transport systems integrating IoT and AI require a comprehensive quality assessment to ensure reliability and effi ciency. The assessment is based on three types of data: objective, indirect and calculated. These data undergo multi-stage processing: normalization, noise fi ltering, identifi cation of key parameters and integration into a single chronological model.The key approaches to reliability assessment are failure-free prediction and residual resource calculation. The model based on the normal distribution of failures is effective in stable conditions, but is limited under dynamic loads. An alternative model links the resource with operational complexity coeffi cients determined by expert methods, which provides fl exibility in conditions of incomplete data. Integration of data, modern algorithms and fl exible models allows increasing the reliability of autonomous transport, reducing maintenance costs. Further development requires international standardization and the introduction of innovative solutions for working in dynamic conditions.