Developing Two-Armed Robot Assistant: Intelligent Tutoring Skills Based System

Abstract

Background/Objectives: The objective of this study is to develop constructive and firmware solutions to create universal robot that would perform different simple tasks and replace human labor in remote autonomous operations. Methods: Specific features of the suggested learning method are represented by the possibilities to copy and exchange skills between different robots including those that possess different kinematic schemes. This becomes possible due to the adaptive control over the manipulator links in line with the current configuration of the effective area. Findings: The study suggests an approach to controlling robots as they perform operations under different conditions including those that are not predetermined by the working area and in cases with obstacles based on earlier obtained knowledge (skills) and according to the universal model of control over performing one particular acquired practice (complex operation with objects) in undetermined environment as a combination of elementary actions performed in real-time mode revealing the capability of adapting to the changing conditions of the environment. Applications/Improvements: The results of the study can be applied to constructive, hardware and software solutions in the process of developing a universal robot to replace manual labor in remote autonomous operations.