DESCRIZIONE
A traffic monitoring device has a monitoring processor with an input interface and an output interface and an AI system which is in bidirectional data communication with the monitoring processor. The AI system has an AI processor, a rule generator based on self-learning algorithms and a reference rule memory. The rule set generator of the AI system is designed to detect patterns and regularities in the traffic movements based on the temporally and spatially resolved observation data received via the input interface of the monitoring processor, and to detect the detected patterns and regularities in the traffic movements as a reference set of rules in the reference set of rules to save. At the request of the monitoring processor, the AI processor is designed to check test data of individual traffic movements received via the input interface for deviations from the reference set of rules stored in the reference set of rules. The monitoring processor is designed to output warning signals via the output interface if the deviations determined by the AI processor are above a predefinable deviation threshold.