Abstract: This article meticulously examines the questions of application of certain technologies of multi-agent systems theory in the area of unmanned traffic management for combatting the so-called “generative adversarial attacks” on the computer vision systems that are used in such vehicles. The article provides examples of generative-adversarial attacks on various types of neural networks, as well as describes the problems that arise when using computer vision. Possible solutions to these problems are proposed. Research methodology includes the theory of multi-agent systems applicable to automobile transport, which suggests using the so-called V2X-interaction, i.e. constant exchange of information between the vehicle and various actors involved in road traffic – a central control system, other vehicles, roadside infrastructure and pedestrians. The authors’ special contribution to this research lies in application of the theory of multi-agent systems for traffic arrangement with consideration of its actors as the agents with diverse roles. The novelty consists in employment of one of the methods of artificial intelligence in solution of the problems, obtained due to the use of other methods of artificial intelligence (recognition of images in computer vision). The relevance of the study is based on the detailed coverage of the questions of organization of unmanned traffic on training grounds and public roads.

Abstract: The subject of research is the architecture of artificial intelligent systems, developed as part of a hybrid approach to artificial intelligence. The article offers the author’s vision of the process of constructing artificial intelligent agents based on a hybrid approach using organismic principles. An artificial intelligent agent with a hybrid scheme is a “cybernetic machine” operating in a certain environment and functionally interacting with it. Of interest is the way the agent interacts and makes decisions, in which information from the environment passes through many sensors, and then it is cleaned up and sensory integrated with further translation into a symbolic form for decision making based on symbolic logic and the operation of a universal output machine. As the main research methodology, a systems engineering approach to the analysis and construction of technical systems was adopted, as well as a functional approach as an additional research method. The novelty of the study is in the use of a hybrid paradigm for constructing artificial intelligent systems in conjunction with systems and functional approaches in the design of technical systems, which made it possible to generalize the available data on the interaction of intelligent agents with the environment and identify interesting patterns for use in the development of artificial intelligence systems. The main conclusion of the study is the possibility of using a hybrid paradigm to obtain artificial intellectual agents that have important advantages of the upward and downward artificial intelligence paradigm - the ability to learn and behave appropriately in an unknown environment and the ability to explain the reasons for their decisions, respectively. This important finding will advance research into explainable artificial intelligence.