Access control system to information resources of swarm robotic systems in the presence of intruded malicious agents

Abstract

A distributed approach to the creation of robotic swarm systems (SRS) will allow solving a wide range of problems in the fields of environmental protection, medicine, cleaning, surveillance, and many others. On the other hand, this approach allows an attacker to intrude malicious agents whose actions aimed at reducing the quality of the entire system. In this regard, the issue of research, design, and testing of these systems from the point of view of information security becomes relevant. Based on this, the purpose of this study is to develop an access control system (ACS) to SRS information resources in the presence of intruded malicious agents to improve the efficiency of SRS functioning when performing spatially distributed tasks. Within the framework of solving the task set, an approach to the synthesis of ACS is proposed, which makes it possible to ensure the continuous detection of malicious agents at all stages of the life cycle of the SRS operation. An element of the scientific novelty of the presented solution is the algorithm for synthesizing an ensemble of heterogeneous classifiers, the main idea of which is a combination of stacking and boosting methods for preparing a training sample and training a meta-classifier. Distinctive features of the proposed algorithm are the ability to use many different classifiers as basic classifiers, as well as the ability to reduce the size of the training set of the meta-classifier by using boosting when selecting samples in the training set. The proposed approach to the synthesis of ACS to SRS information resources using the presented algorithm improves the accuracy of detecting malicious agents both in the process of task distribution and in the process of task execution. The presented solution is implemented in the form of software in the Python programming language, which can be used in modeling decentralized control systems of SRS. The article refers to the scientific specialty 2.3.6 “Methods and systems of information protection, information security”.