Yur'eva R.A., Komarov I.I., Maslennikov O.S. Development of the method for detection and identification of hidden destructive impact on the multi-agent robotic systems

Abstract: Increased information security risks in multi-agent robotic systems create a need for new and known assessment in terms of security algorithms. Such risks include the loss or inaccessibility of data, spreading false information about the purpose of grouping and the use of distorted information, lack of energy resources. Authors note that common approaches to information security in multi-agent robotic systems so far are not formed. The research is aimed at developing a model of information security in multi-agent robotic systems taking into account the specifics of the technology. Existing scientific and methodical apparatus and technical solutions of ensuring information security in multi-agent systems are not applicable to the tasks of ensuring information security in multi-agent robotic systems due to the specific technology and a special type of threat models and the offending patterns associated with them. Existing methods of providing information security of multi-agent systems do not provide a comprehensive solution of the problems of information security in the multi-agent robotic systems, since they do not take into account the specificity of their composition and structure. Scientific novelty consists in the development of software models of information security risks in multi-agent robotic systems. An advantage of the method is the ability to detect new types of attacks without having to modify or update the model parameters, since the invasion of the offending system can be described as a deviation from the nominal behavior.

Keywords: