Nowadays, the Internet of Things (IoT) has been growing continuously along with different networks providing connectivity methods for IoT devices. In this context, LoRa networks, or LoRa mesh networks, have been developed for long-range, low-power communication for sensors. This paper studies an application of image transmission over the LoRa mesh network by considering different images and energy consumption during a series of experiments. The experimental results show a high packet delivery ratio via the network (over 90%). In addition, the experiments also show the possibility of transferring 500 images taken by battery-powered ESP32-CAM with a frame size of 480 × 320 and a quality level of 30.

The concept of a Smart City is to use the existing resources in an optimal way to provide the greatest convenience to its residents. This requires close integration of all components, for example, street video surveillance, public services, intelligent transport systems and others, on the scale of a megalopolis. Every year, the world’s megacities are becoming more comfortable for residents due to the introduction of newest technologies. First, such technologies include intelligent control systems in the transportation field. The main goals of Smart Transportation are the efficient and coordinated movement of people, monitoring the location of objects, fast and reliable interaction of vehicles with each other, as well as guaranteeing road safety. This paper represents examples of artificial intelligence technologies and optimization methods applications to create such smart systems.

The paper proposes and tested an approach to the video data formation processing pipeline that solves the problem of automating the control of the presence and correctness of wearing personal protective equipment by personnel in difficult conditions of filming by CCTV cameras. The proposed solution is based on neural network algorithms and is flexibly configured for specific conditions and tasks, including when generating operational alerts. The approach is demonstrated on the example of recognition of medical masks. The solution is based on: an automated markup system for training a detector based on AlphaPose, a YOLOX neural network detector, a ByteTrack tracking system, and a classifier based on a lightweight CNN, the input of which is minitracks of people’s faces, 8 frames each. The collected dataset consists of over 260,000 minitracks. Achieved the quality of classification $F_1=0.86$ to determine the presence and $F_1=0.79$ to determine the correct wearing of the mask.

The article examines the problem of ensuring the political stability of a democratic social system with a shortage of a vital commodity (like vaccine against COVID-19). In such a system, members of society citizens assess the authorities. Thus, actions by the authorities to increase the supply of this commodity can contribute to citizens' approval and hence political stability. However, this supply is influenced by random factors, the actions of competitors, etc. Therefore, citizens do not have sufficient information about all the possibilities of supplying, and it is difficult for them to make the right decisions. Such citizen unawareness can be exploited by unscrupulous politicians to achieve personal targets. Therefore, it is neces¬sary to organize public control to motivate politicians to use all available opportunities in supplying. The goal of the paper is to build such a digital mechanism of public control of the politicians by citizens, which would best assess and stimulate the activities of the authorities to improve the supply of a vital commodity. In the age of artificial intelligence, such digital public control in the face of uncertainty can be based on digital machine learning. In addition, it is necessary to take into account and model the activities of politicians associated with the presence of their own targets that do not coincide with public ones. Such politicians can use the learning of citizens for their own targets. The objective of the article is to build an optimal digital mechanism of public control in a two-level model of a democratic social system—a digital society. At its top level, there is the Citizen, who gives an assessment for the Politico located at the lower level. In turn, the Politico can influ¬ence the supplying of a vital commodity. Political stability is guaranteed if the Citizen regularly approves of the Politico’s actions to increase this supply. However, the Politico may not use the opportunities available to him to offer a commodity to achieve a personal target. To avoid this, the Politico’s control mechanism is proposed. It includes the procedure for digital learning of the Citizen, as well as a procedure for assessing the Politico activity. Sufficient conditions have been found for the synthesis of such the Politico’s control mechanism, at which stochastic possibilities of increasing the supply of a vital commodity are used. The example of such the Politico’s control mechanism is considered on the case of supply of the COVID-19 vaccine in England.

The problem of stabilizing the chain of three integrators by a piecewise continuous constrained control is studied. A feedback law in the form of three nested saturators specified by six—three model and three design—parameters is proposed. Global stability of the closed-loop system is studied, and an optimization problem of determining the feedback coefficients ensuring the greatest convergence rate near the equilibrium while preserving global asymptotic stability is stated. It is shown that the loss of global stability results from arising hidden attractors, which come to existence when the convergence rate becomes greater than or equal to a critical value depending on the control resource. A numerical procedure for constructing hidden attractors is developed. The bifurcation value of the convergence rate, which is an exact upper bound of the parameter values ensuring global asymptotic stability of the closed-loop system, is determined numerically by solving an algebraic system of four equations.