A METHOD FOR ASSESSING THE INFLUENCE OF GEOMETRIC IMAGE DISTORTIONS ON THE RESULTS OF THEIR COMPARISON WITHIN THE CORRELATION-EXTREME ALGORITHM

The article presents a comprehensive study of methods for quantifying the potential accuracy of determining the coordinates of various types of objects using matrix correlation-extreme navigation systems. Three main categories of objects are considered: point objects (occupying a small part of the frame), extended objects, and planar objects (occupying a significant part of the frame), which can be either tangent or non-tangent to the observation system. Particular attention is paid to the analysis of factors that affect the accuracy of measuring the coordinates of these objects. These include geometrical parameters of objects, their spatial orientation, contrast rela- tive to the background, lighting conditions, atmospheric phenomena, and technical characteristics of matrix sensors. Mathematical models describing the relationship between these factors and the potential measurement accuracy are presented. The peculiarities of step-by-step object navigation using matrix correlation-extreme navigation systems are investigated, which can significantly improve the accuracy of coordinate determination. The algorithms for optimising the process of sighting, accounting for the specifics of different types of objects, are proposed. The methods of compensation for systematic errors and minimisation of the influence of random interfer- ence are considered. The results of experimental studies confirming the effectiveness of the proposed methods of accuracy assessment are presented. It is established that with the optimal choice of system parameters and information processing algorithms, it is possible to achieve high accuracy in determining the coordinates of all types of objects under study. The developed methods and algorithms can be effectively used in the design and configuration of matrix correlation-extreme navigation systems for various practical applications, including navigation systems for robotic systems and other moving objects, where high accuracy of coordinate determination relative to ground reference points is required. The obtained results create a theoretical basis for further improvement of methods for increasing the accuracy of matrix correlation-extreme navigation systems and expanding the scope of their practical application.