Using Patterns for Autonomous Mobile Robot Control Tasks

The issue of increasing the accuracy of performing control tasks for an autonomous mobile robot (AMR) by applying patterns to its structural elements is considered. The functions of comparing patterns with the intended purpose of determining the degree of their similarity with the standards are considered. We will call a pattern a special label that is applied to the structural elements of an AMR and projected into the image of the onboard vision system (oVS). The pattern may represent a combination of black and white primitive figures of a unique shape applied in a limited area of the surface of the mobile robot structure. The purpose of the research was to study ways to improve the reliability of controlling an AMR using patterns and an oVS. Research corresponds to the current direction of development of autonomous robotic systems for a wide range of applications. Two problems solved with the help of the pattern-vision system are considered. The first is the monitoring of the parameters of the oVS of an autonomous mobile robot. The second task is to improve the accuracy of manipulator control. Both tasks are interrelated, since without knowing the exact options of the technical vision system, it is impossible to accurately control the manipulators of an AMR. In both cases, the pattern comparison function is calculated in order to determine the degree of their similarity with the standards. Pattern standards can be stored in the memory of the onboard computer system, for example, for the task of monitoring the options of the technical vision system itself or calculated for the task of manipulator control. The function of comparing a pattern with a standard should reflect the degree of their similarity and be computationally simple. From the point of view of the criterion of computational simplicity, three functions for comparing a pattern with a standard are considered.