Detecting abnormal events in vehicle operation based on machine learning analysis of messages transmitted over communication channels

Abstract

A computer implemented method of identifying an abnormal event in an operational environment of a vehicle, comprising using one or more processors adapted for receiving a plurality of messages intercepted by one or more devices adapted to monitor messages transmitted via one or more segments of one or more communication channels of a vehicle, applying a plurality of trained machine learning models to evaluate compliance of each of the intercepted messages with one or more baseline models according to feature(s) identified for each intercepted message, the baseline model(s) defining learned message patterns is created by training the plurality of machine learning models with a plurality of training datasets comprising training messages reflecting valid operation of the vehicle, identifying incompliant intercepted message(s) which is incompliant with the baseline model(s), the incompliant message(s) is transmitted as result of one or more abnormal event and generating an alert indicative of the abnormal event(s).