DEVELOPMENT PROSPECTS OF LEARNING-BASED CONTROL METHODS FOR ROBOT MANIPULATORS

Abstract

Learning-based control methods for robot manipulators have been developed over the past decade. In the future, this direction is expected to become one of the most promising approaches in the field of robot manipulator control. In robot manipulators, learning-based control is mainly applied to address problems such as friction forces arising in the joints of mechanisms and other uncertainties that may exist in dynamic models. Such factors significantly complicate the system and sometimes make it impossible to model them accurately using mathematical methods. This article analyzes learning-based control approaches in robotic manipulators. It also explains some of the existing challenges in this field. The presented review may serve as a general direction and methodological roadmap for future research in the area of learning control for robotic manipulators.