Application of GNSS systems for automated control of working processes of graders with artificial intelligence

Abstract

Problem. Despite rather successful application of artificial intelligence to control graders, these systems have a number of disadvantages associated with their use: complexity of choosing the system architecture; batch learning and multi-epoch learning require significant time resources; many of the existing computational intelligence systems cannot determine the evolving rules by which the system develops; problems when considering a multitude of indicators that have a complex structure of relationships and contradict each other; difficulty of taking into account the indirect influence of interdependent components in conditions of uncertainty; non-linear character of mutual influence of objects and processes, non-stochastic uncertainty, non-linearity of mutual influence, partial inconsistency and significant interdependence of components. Goal. There is an urgent scientific task of developing a methodology for creating intelligent road vehicle control systems using artificial neural networks and fuzzy cognitive models. Methodology. The methodology of assessment and forecasting in intelligent decision support systems consists of the following sequence of actions: 1) input of initial data. At this stage, the initial data available on the object to be analyzed are entered. Initialization of the basic object state model is carried out; 2) identification of factors and connections between them. Analysis of the models of multi-criteria evaluation of alternatives under conditions of uncertainty showed that the values of model parameters are often represented by intervals, as there are differences of opinion when obtaining parameter values. When there is interval and fuzzy information, it is advisable to use the fuzzy-interval method. Results. This paper briefly examines the features of using neural network technologies, machine control systems and training to increase the productivity of graders. A formalized description of the task of analyzing and forecasting the state of objects in intelligent decision-making support systems of the intelligent system of the road vehicle is carried out. The specified formalization allows us to describe the processes that take place in intelligent decision support systems when solving the tasks of analyzing and forecasting the state of objects of the road machine intelligent system. Originality. An overview of intelligent control systems that can be used for automated control of work processes of earthmoving machines with the use of artificial intelligence is presented, a methodology for implementing practical tasks is proposed. Practical value. An overview of intelligent control systems is presented, practical implementation in programming languages is proposed, the influence of ground conditions on the process of automated work is analyzed, the concept of assessment and forecasting methods in intelligent decision support systems for an intelligent system of graders is given.