Proactive management of the composition and structure of the spatial monitoring system under the influence of destabilizing factors

The article considers issues of increasing the target performance of the spatial monitoring system expressed in the maximization of the number of detected search objects under the influence of destabilizing factors. It is shown that such a task can be solved by integrating simulation modeling and artificial intelligence in the modeling and forecasting module of the proactive management subsystem of the spatial monitoring system. Within the framework of the proposed approach, the simulation model allows generating many variants of the course of an antagonistic conflict and is used as a training ground for the neural network module, which is responsible for the structure of the spatial monitoring system. The interaction between the neural network module and the simulation model is realized by integrating a mental agent into the simulation model with reinforcement learning. It is revealed that the proposed integration is possible by applying an agent-based approach. Based on this approach, the paper presents a structural and functional description of the simulation model for the spatial monitoring system that functions under the influence of destabilizing factors. The results of simulation modeling, which confirm the effectiveness of recommendations to manage the elements of the system using the neural network module trained during the simulation, are also presented. The comparison with the basic strategy of the object search is executed. The authors outline the prospects of applying the neural network technology and reinforcement machine learning in the proactive control subsystem of the spatial monitoring system and ways to achieve them.