COMPARATIVE ANALYSIS OF THE DEGREE OF ROBOTICS COMPLEX AUTONOMY

Abstract

Against the background of the ever-increasing needs for robotic complexes with an increased degree of autonomy and the planned transition to their widespread use, the need for technologies for assessing the quality and comparing the degree of autonomy of such devices is being actualized. The article describes the current state of issues of evaluation and comparison of the degree of autonomy of unmanned complexes. Known estimates of the degree of autonomy are given. In the existing classification system, informational and intellectual autonomy are distinguished, which are considered in close connection. Solutions are proposed that complement the known approaches to the general definition of the degree of autonomy and allow us to form quantitative estimates of the degree of autonomy of robots in various areas of the national economy. Technologies aimed at automating the receipt of these estimates are considered. In particular, the possibility of using the tools of fuzzy cognitive maps well mastered by domestic researchers to determine the degree of autonomy in conditions of incomplete information, the availability of high-quality information and the influence of the human factor is discussed. The necessity of developing ontologies of subject areas is substantiated in order to ensure the possibility of comparing the degree of autonomy of various robotic complexes (RC) and their groupings. In general, the approach is indicated, which is aimed at systematizing assessments of the quality and effectiveness of the use of autonomous robots, and can allow in a short time to prepare a methodological basis for the widespread introduction of robotics. One of the positive consequences of such a systematic approach is the unification of formulations and solutions (modules) in the tasks of RC information support, which, in turn, facilitates interaction between users, customers and developers. For RС developers, a systematic approach makes it possible to reuse successful solutions in various combinations. In conclusion, wishes are expressed to the community of domestic roboticists in joining efforts to unify terminology, describe problem statements and metrics of intelligence of robotic complexes