Development of Methods for Monitoring and Predicting the Safety of Automated Vehicles in Operation

The article studies technologies of storage, processing, remote high-speed transmission, analysis and verification of large amounts of data suitable for use in the field of automotive telematics. The methods of collection, storage, structuring and maintenance of statistics of automated vehicles operation data are described. The logic of structuring and archiving of data streams coming from the server is given. Diagnostics and forecasting of the technical condition of automated vehicles is proposed to be carried out on the basis of data on errors in the standard systems of vehicles, data on failures and failures in the systems included in the automated driving system and on the basis of dynamic control parameters. The developed algorithm for predicting the risks of failures and malfunctions of units and systems affecting safety on the basis of operation data is presented. The approach to identification and forecasting of incorrect parameters of automated vehicles functioning affecting safety is proposed, in particular, the algorithm of analyzing the delay time of the control signal on steering of the automated vehicle on the basis of operation data is developed. The adequacy of the developed algorithm has been verified using the operation data of automated vehicles on public roads, the results of which allow us to assert that this method can be effectively applied to predict the technical condition of automated vehicles.