Manipulator's Control System with Application of the Machine Learning

Abstract

This paper provides an analytical review of the subject area, approaches and methods for real-time control systems with application of the machine learning. The review helps to choose approaches to control robotic devices, methods of artificial intelligence to perform pattern recognition and classification, technologies to design a control system and simulate a robotic device. The design and software implementation stage involved the creation of a structural diagram and description of the selected technologies, training a neural network for recognition and classification of geometric objects, software implementation of control system components. The control system was implemented using the Swift programming language and other technologies, and the configuration and creation of a neural network using the CreateML framework.