The paper considers three technologies for obtaining data on the road surface -through video recording, thermal imaging and laser scanning for the purpose of monitoring, diagnostics and control of the road quality. An analysis of the first two methods showed their significant drawbacks, such as the inability to measure the geometric parameters of deformations (video recording) and the significant dependence of the measurement results on external conditions (thermal imaging). Laser scanning, on the contrary, has a number of advantages, including coordinate referencing, obtaining a three-dimensional model, its transformation and measurement of parameters. Laser scanning is widely used, but mainly for measuring the quantitative characteristics of objects. The paper discusses the application of the laser scanning method to determine the qualitative characteristics of the road surface - the presence or absence of defects, which include hollow spots, waves, cavities, chipping, bleeding, humps, cracks, vertical displacement of road plates, rutting, unevenness of patching, damage to the road surface, track, breach, destruction of the pavement edge, subsidence followed by a complex of repair work. For this, a ground-based laser scanning was performed, the results of which were processed using the Leica Cyclone 9.4 software. According to the scanning data, defects were detected in the form of soil subsidence, hollow spots and humps. The performed work revealed a drawback of the laser scanning method, which consists in the absence of automated detection and recognition of deformations. A number of measures have been proposed to improve this drawback, which slows down the randomness and quality of work in monitoring and diagnosing the road. Further prospects for research on this topic, in particular the multi-purpose use of scanning data, by creating a distributed ledger are also indicated.

The paper is devoted to topical issues of monitoring infrastructure facilities. The aim of the study is to compare the already established traditional method of ground-based laser scanning and its aerial analogue based on the use of compact drones in collecting data on road structures. The question was raised about the possible ways to modernize the existing monitoring procedure for the timely detection and prevention of dangerous emergencies at engineering structures. The most effective technologies for obtaining comprehensive information on the technical condition of such structures are analyzed. It was determined that the most rational method from the point of view of information content, mobility and examination time is the method of using airborne laser scanners in conjunction with unmanned small-sized aircraft. The procedure of the entire cycle of obtaining information, including its subsequent processing, is described. The analysis of stationary and other applied methods of monitoring infrastructure facilities is made, their advantages and disadvantages are described. A question was raised about the prospect of creating software that uses artificial neural networks as a means of automating the processes of subsequent processing of primary scan data and analysis of the obtained object model. Conclusions are made about the appropriateness and prospects of using similar tools and methods for the needs of monitoring engineering infrastructure facilities, as well as the need for further development of software that will automatically analyze the accumulated data on the same infrastructure object.

The paper discusses the task of evaluating the possibility of using robotic systems (intelligent agents) as a way to solve a problem of monitoring complex infrastructure objects, such as buildings, structures, bridges, roads and other transport infrastructure objects. Methods and algorithms for implementing behavioral strategies of robots, in particular, search algorithms based on decision trees, are examined. The emphasis is placed on the importance of forming the ability of robots to self-learn through reinforcement learning associated with modeling the behavior of living creatures when interacting with unknown elements of the environment. The Q-learning method is considered as one of the types of reinforcement learning that introduces the concept of action value, as well as the approach of “hierarchical reinforcement learning” and its varieties “Options Framework”, “Feudal”, “MaxQ”. The problems of determining such parameters as the value and reward function of agents (mobile robots), as well as the mandatory presence of a subsystem of technical vision, are identified in the segmentation of macro actions. Thus, the implementation of the task of segmentation of macro-actions requires improving the methodological base by applying intelligent algorithms and methods, including deep clustering methods. Improving the effectiveness of hierarchical training with reinforcement when mobile robots operate in conditions of lack of information about the monitoring object is possible by transmitting visual information in a variety of states, which will also increase the portability of experience between them in the future when performing tasks on various objects.

A centralized multiagent system based on the methods of feudal reinforcement learning, including agents-managers and agents-subordinates, is considered. A brief review of the author’s previous works on this topic is given. The standard algorithm for the functioning of systems of this type is considered, including the translation of the decision maker to agents-managers, the division of tasks by agents-managers into a set of subtasks, the choice by the agent-manager of the strategy used, the formation of reward functions by agents-managers, the assignment of tasks to agents-subordinates, the execution by agents-subordinates assigned tasks. The main problems of this algorithm are presented, changes are made to ensure the possibility of automatically assigning agent-managers and forming groups of subordinate agents around them, reproducing and exchanging experience. More attention is paid to the problem of experience exchange, the main ways of experience exchange are given. The principles of operation of a machine vision system that implements an upgraded algorithm are described. An assessment of the effectiveness of the obtained algorithm for the collective interaction of intelligent agents using a software model developed in Microsoft Unity is given. A comparison is made between the standard algorithm for multiagent interaction and the proposed algorithm for the collective interaction of intelligent agents in centralized multi-agent systems based on the approach of reinforcement learning and visualization of three-dimensional scenes. The conclusion is made about the expediency of using the developed algorithmt.

The system analysis of the hierarchical intelligent multi-agent system in general, as well as its main structural unit, the intelligent agent, its major subsystems identified. As part of the analysis of the computer vision subsystem, it was concluded that the considered sources have insufficiently worked out issues related to the processing of occlusions, with the automation of the process of reconstruction of three-dimensional scenes, with the implementation of the processing of an unstructured set of images. The structure of the block for the reconstruction of three-dimensional scenes is proposed, the implementation of which is aimed at eliminating the indicated problems characteristic of the machine vision subsystem. The analysis of the main methods of implementing unsupervised learning is carried out, based on the results of which it is concluded that it is advisable to use reinforcement learning when implementing systems of this type. Such types of reinforcement learning as hierarchical reinforcement learning and multi-agent reinforcement learning are considered. A method for segmentation of macro actions is proposed, based on the implementation of clustering by the method of label propagation, in which the number of transitions is formalized in the form of weight coefficients of edges.