The use of optical markers in the angular position control system of objects

Abstract

Relevance and goals: Pattern recognition systems are increasingly used in modern technology. The development of the theoretical base of image processing and the widespread use of free open source libraries make it possible to use new solutions in a variety of applied problems. One of these tasks is to determine the angular position of the object in the study of its radar response. A feature of the task is the impossibility of direct control of the position of the object through the mechanism of its drive, and the use of additional position sensors is also unacceptable. The purpose of this study is to determine the possibility of practical use of optical markers for solving the problem, as well as assessing the quality characteristics of recognizing markers of various types.Materials and methods: The solution of the problem was achieved using the methods of controlling the position of an object by optical markers by analyzing the image of the object. At the same time, the functions of the OpenCV machine vision library adapted to the task and the general elements of the machine vision theory were used. Also, a hardware platform was chosen that provides autonomous device execution, taking into account the required performance.Results: In the course of the work, a choice of suitable types of optical markers was made and qualitative characteristics were obtained for determining the coordinates of markers in different observation conditions. The library functions were adapted to work with the selected markers. The analysis of the speed characteristics of the system when working on the selected hardware platform. Findings: Studies have shown the possibility of practical use of optical markers for solving the problem, and also allowed to choose the most suitable type of markers. The selected hardware and software platform provides the ability to solve problems in real time.