Special features of using neuronetwork technolo-ies to control the work processes of bulldozers with a gnss-intensifier

Abstract

Abstract. Problem. The specificity of modern road construction works requires evaluation of the situation; forecast of the machine's normal operation and the development of emergency situations; synthesis and evaluation of possible actions of the operator and selection of the best ones. That is, there is a problem of effective support of the road car, whose solution is transferred to the intelligence of the car itself. A distinctive feature of intelligent systems is the ability to plan behaviour, adapt and learn. An intelligent system automatically changes the algorithms of its functioning and the structure in order to preserve or achieve an optimal state when external conditions change. The integration of algorithmic methods of controlling complex objects and methods of artificial intelligence is aimed at solving the tasks with uncertainty of the source information. The implementation of intelligent systems shows their prospects in the field of road construction works. Thus, optimization of road machine work processes, increasing the efficiency of road-construction machines (RCM) due to an intelligent control system is an urgent task. Studying the structure of the intelligent system of the road machine is relevant as well as developing a mathematical model of intelligent optimization of work processes based on the complex of readings of integrated GNSS sensors. Goal. The purpose of the research is to increase the efficiency of construction machines through their intellectualization, development of scientific algorithms for creating mathematical models of the working bodies of the RCM with SNS-intensifier of work processes of interaction with the soil environment. Methodology. In the paper the scientific methods of system analysis and decision-making theory for the structuring and synthesis of intelligent control systems are used; mathematical modeling methods for building and researching hydraulic and dynamic models of road vehicles; theories of automatic control under conditions of uncertainty; methods of object-oriented programming for simulation modeling of a road machine work process during operations. Results. For the first time, the methods of solving the problem of controlling the hydraulic drive of a road machine were offered, which, unlike traditional ones, take into account the execution of digging operations in conditions of a short decision-making time, which ensures the movement of the working body along an arbitrary trajectory in a short time with high accuracy. Originality. This paper is one of the first in Ukraine in which the scientific problem of increasing the efficiency of RCM at the interaction of working bodies with the soil environment in conditions of limited time for decision-making was formulated and solved. Practical value. Time of work is cut due to the reduced number of passes during soil development and increasing the productivity of RCM, which will have a positive effect on financial and labour costs, ensuring the possibility of continuous operation, which increases the productivity of RCM.