An adaptive model for detecting the vulnerabilities of unmanned vehicles interface in smart city infrastructure

A method of detection vulnerabilities in unmanned vehicle interfaces based on an analysis of a traffic condition in communication channels of unmanned transport systems is proposed. The approach is based on non-parametric statistics methods for assessing the information states of controlled objects, which include such unmanned vehicles resources as: communication channel, processor, memory, power supply, etc. For each of these resources, it is proposed to evaluate the change in such characteristics as the degree of load resource and its rate of change. Recognition of the state of network traffic is carried out in conditions of a lack of a priori information about the properties of the intrusion source and the stochastic nature of the recognized events. To increase the reliability level of vulnerability detection in the model, adaptive dynamic tuning of decision-making rules for classifying the information state of the traffic of unmanned vehicles is carried out.